From 4a2fe7882197d6f49197ffde224224d93432105d Mon Sep 17 00:00:00 2001 From: wantsong Date: Wed, 24 Jun 2026 07:02:09 +0800 Subject: [PATCH] =?UTF-8?q?=E5=A2=9E=E5=8A=A0=E4=BA=BA=E7=89=A9=E4=B8=80?= =?UTF-8?q?=E8=87=B4=E6=80=A7=E5=8D=95=E9=95=9C=E9=AA=A8=E6=9E=B6?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- VIDEO_WORKBENCH.md | 41 + .../fix-title-batch-report.md | 11 + .../阶段五-人物一致性单镜.fixed.md | 1059 +++++++++++++++++ .../阶段五-人物一致性单镜.heading-map.json | 516 ++++++++ .../阶段五-人物一致性单镜.heading-plan.json | 618 ++++++++++ .../阶段五-人物一致性单镜.heading-report.md | 121 ++ .../intake/阶段五-人物一致性单镜.md | 1059 +++++++++++++++++ 7 files changed, 3425 insertions(+) create mode 100644 projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/fix-title-batch-report.md create mode 100644 projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.fixed.md create mode 100644 projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-map.json create mode 100644 projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-plan.json create mode 100644 projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-report.md create mode 100644 projects/2026-06-23-在路上/intake/阶段五-人物一致性单镜.md diff --git a/VIDEO_WORKBENCH.md b/VIDEO_WORKBENCH.md index 179ccd6..dae69b0 100644 --- a/VIDEO_WORKBENCH.md +++ b/VIDEO_WORKBENCH.md @@ -92,6 +92,47 @@ It should not repeat the project ID, full intake file index, or project map from File naming can be agreed during the first real project. Do not create a rigid naming scheme before the real GPT outputs are known. +### Intake Heading Repair + +When copied GPT / LLM intake Markdown has broken heading hierarchy, use the installed `fix-title` Skill as an external repair capability, but keep the calling protocol local to Video Workbench. + +Local calling rules: + +- Repair work must run through a real Agent / subagent invocation when the runtime provides one. +- The main session should pass the source file paths, `mode=artifact` or `mode=discussion`, the `fix-title` skill path, and the output directory to the worker. +- The main session should not build semantic heading plans or apply heading edits itself when an Agent carrier is available. +- For files under `projects//intake/`, output must go under: + +```text +projects//intake-repair/fix-title-YYYY-MM-DD-/ +``` + +The repair output package should contain: + +```text +.heading-map.json +.heading-plan.json +.fixed.md +.heading-report.md +fix-title-batch-report.md +``` + +If repaired files replace originals, first copy the original Markdown files to: + +```text +projects//intake-repair/fix-title-YYYY-MM-DD-/originals-before-replacement/ +``` + +Main-session responsibilities after the worker returns: + +- verify the returned files exist; +- inspect the heading map, plan, fixed copy, per-file report, and batch report; +- check that intended artifact files have one clear document title unless the report flags an intentional appended artifact; +- check that ambiguous appended artifacts are reported instead of silently merged; +- decide whether to replace originals, and write replacement notes when replacement happens. + +Do not put durable intake repair output under `tmp/`. Keep reusable automation changes in `skills-vault`; Video Workbench owns only the local invocation protocol, project evidence, replacement notes, and supplier-request records. + ## 7. Visual System Visual projects use: diff --git a/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/fix-title-batch-report.md b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/fix-title-batch-report.md new file mode 100644 index 0000000..3e21f5b --- /dev/null +++ b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/fix-title-batch-report.md @@ -0,0 +1,11 @@ +# Fix Title Batch Report + +| Source | Mode | Heading Map | Plan | Fixed | Report | +| --- | --- | --- | --- | --- | --- | +| C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake\阶段五-人物一致性单镜.md | artifact | C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake-repair\fix-title-2026-06-24-character-consistency\阶段五-人物一致性单镜.heading-map.json | C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake-repair\fix-title-2026-06-24-character-consistency\阶段五-人物一致性单镜.heading-plan.json | C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake-repair\fix-title-2026-06-24-character-consistency\阶段五-人物一致性单镜.fixed.md | C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake-repair\fix-title-2026-06-24-character-consistency\阶段五-人物一致性单镜.heading-report.md | + +## Run Summary + +- Files processed: 1 +- Changed headings: 101 +- Ambiguities: none detected diff --git a/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.fixed.md b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.fixed.md new file mode 100644 index 0000000..4db64cd --- /dev/null +++ b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.fixed.md @@ -0,0 +1,1059 @@ +# 《在路上》人物一致性验证单镜骨架 V1 + +本次展开 5 个镜头: + +* **S-05:斑驳行囊** +* **S-14:疲惫的人 / 你醒了吗** +* **S-21:那一簇火,指引方向** +* **S-25:我擦拭微尘** +* **S-33:我咬紧牙关** + +这 5 个镜头的作用,不只是服务歌词,而是共同验证一个问题: + +> 在背影、侧背、半侧脸、手部动作、情绪近点里,主角是不是仍然像同一个“路上行者”。 + +--- + +## 0. 这 5 镜的人物一致性验证分工 + +| 镜头 | 验证重点 | 人物可见度 | 主要锚点 | +| ---- | -------- | --------- | ------------ | +| S-05 | 背影与行囊一致性 | 背影局部 / 中近 | 背包、外套下摆、肩背轮廓 | +| S-21 | 侧背动态一致性 | 中远景 / 侧背影 | 调整背包、火光、继续前行 | +| S-14 | 半侧脸气质一致性 | 半身偏侧 | 年龄、胡茬、疲惫但不塌 | +| S-25 | 手部动作与微尘感 | 局部 / 半侧 | 手、脸侧、外套、风尘感 | +| S-33 | 情绪近点一致性 | 半侧脸短近景 | 牙关、胡茬、眼神有火 | + +测试顺序建议: + +> **S-05 → S-21 → S-14 → S-25 → S-33** + +先从背影和服装稳定性开始,再逐步靠近人物脸部,避免一上来陷入“像不像证件照”的判断。 + +--- + +## 镜头 S-05:斑驳行囊 + +### 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 斑驳行囊 + +时间位置约: + +> **0:16 - 0:32** + +这是早段人物锚点第一次被观众看清的镜头。 + +--- + +### 2. 本镜情绪功能 + +这个镜头不负责表现人物脸。 +它负责告诉观众: + +> 这个人在路上已经走了很久。 +> 不是刚出发,不是摆拍,而是真的背着一段时间和重量在走。 + +它也是后续所有背影、远景小人物镜头的服装和道具基准。 + +--- + +### 3. 核心画面 + +清晨冷光里,主角背对镜头行走。 + +画面重点不是脸,而是: + +* 旧背包; +* 深灰 / 暗卡其长外套; +* 外套下摆被风轻轻掀起; +* 背包肩带压在肩上; +* 靴子偶尔进入画面下缘; +* 泥泞路面和风沙背景轻微可见。 + +人物可以是中近的背影局部,也可以是中远景背影。 +但不要拍成“背影写真”。 + +--- + +### 4. 人物与镜头关系 + +#### 人物可见度 + +* 背影; +* 不露脸; +* 重点在肩背、背包、外套轮廓; +* 靴子可以部分出现。 + +#### 人物占比 + +建议人物占画面: + +> **1/4 左右以内** + +因为这镜是道具锚点,可以比普通远景稍近,但仍不能变成人物写真。 + +--- + +### 5. 人物一致性锚点 + +这一镜必须锁定 5 个稳定元素: + +1. **背包形状** + 旧、实用、体量适中,不像潮流机能包。 + +2. **外套长度** + 长外套下摆要形成背影识别特征。 + +3. **肩背姿态** + 有重量感,肩膀略沉,但不能塌。 + +4. **发型后轮廓** + 深色、稍乱,中年感,不要潮男造型。 + +5. **整体气质** + 风尘、疲惫、继续走,不是旅行博主。 + +--- + +### 6. 场景与视觉母题 + +调用: + +* 清晨薄霜; +* 斑驳行囊; +* 旧背包; +* 长外套; +* 泥路; +* 风。 + +这是人物锚点第一次明确出现,应该和后续人物锚点板保持高度一致。 + +--- + +### 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 背影偏画面一侧; +* 路和环境仍要占一定比例; +* 背包和外套处于视觉中心; +* 不要正面,不要回头。 + +画面应该像: + +> 镜头从后方跟了一小段路,观众第一次注意到这个人背上的重量。 + +--- + +### 8. 光线与色彩 + +* 清晨冷光; +* 低饱和; +* 灰白薄霜; +* 土褐泥路; +* 外套深灰 / 暗卡其; +* 背包旧而暗。 + +不要暖黄旅行片光线。 +不要把背包拍得太新、太干净、太户外广告。 + +--- + +### 9. 与前后镜关系 + +#### 前一镜 S-04 + +清晨薄霜,世界先出现。 + +#### 本镜 S-05 + +人物锚点出现,但只给背影和行囊。 + +#### 后一镜 S-06 + +主角站在岔路前,进入“你要去哪”的方向问题。 + +--- + +### 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* 背影; +* 旧背包; +* 长外套; +* 清晨泥路; +* 行走中; +* 不看镜头; +* 不要肖像感; +* 不要旅行广告感。 + +这不是最终 Prompt,只是执行要点。 + +--- + +### 11. 验收关注点 + +审这一镜时看 6 件事: + +1. 背包是否稳定、可复用? +2. 外套下摆是否形成识别轮廓? +3. 背影是否像“同一个行者”? +4. 有没有旅行写真感? +5. 人物有没有太精致? +6. 这张图缩小后,还能不能认出是这个角色? + +--- + +## 镜头 S-14:疲惫的人 / 你醒了吗 + +### 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 疲惫的人 +> 你醒了吗 + +时间位置约: + +> **0:32 - 1:05 段落末尾** + +这是城市压迫段里的一个人物确认镜头。 + +--- + +### 2. 本镜情绪功能 + +这一镜是全片早段少数可以接近人物的镜头。 +它要完成的是: + +> 让观众第一次确认:这个行者是一个 50 岁左右的中年人,疲惫,但精神没有塌。 + +它不是人物写真。 +不是“拍一个沧桑男人很帅”。 +它是让观众在城市阴影里看见一个被现实磨过的人。 + +--- + +### 3. 核心画面 + +城市边缘阴影里,主角短暂停顿。 + +画面里可以有: + +* 半侧脸; +* 胡茬; +* 被风吹乱的深色头发; +* 深色长外套领口; +* 旧背包肩带; +* 背后是冷硬混凝土、钢铁结构或湿冷墙面。 + +人物不需要直视镜头。 +最好是看向画面外侧或低头后微微抬眼。 + +--- + +### 4. 人物与镜头关系 + +#### 人物可见度 + +* 半身偏侧; +* 半侧脸; +* 不完全正脸; +* 不直视镜头或只短暂擦过镜头方向。 + +#### 人物占比 + +可以比多数镜头稍大: + +> **1/3 左右以内** + +但仍要保留城市阴影和环境压迫感。 + +--- + +### 5. 人物一致性锚点 + +这一镜验证的是脸部和气质: + +1. **年龄感** + 50 岁左右,不要年轻化。 + +2. **胡茬** + 明显,但不能修成潮流胡型。 + +3. **头发** + 深色,稍乱,有风尘感,不要精致发型。 + +4. **眼神** + 疲惫,但不能空掉;有一点火。 + +5. **气质边界** + 不能像流浪汉,不能像时尚男模,不能像电影硬汉摆拍。 + +--- + +### 6. 场景与视觉母题 + +调用: + +* 城市边缘; +* 钢铁峡谷后的阴影; +* 疲惫的人; +* 中年行者; +* 冷灰色光线。 + +这一镜是 S-10 钢铁峡谷之后的情绪落点: + +> 城市把人压小之后,观众短暂看见这个人的脸。 + +--- + +### 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 人物偏侧,不居中; +* 背后有冷硬城市结构; +* 半侧脸处于暗部和冷光之间; +* 镜头不要太近; +* 不要海报式人像构图。 + +这一镜的关键不是“好看”,而是: + +> 疲惫是真的,尊严也是真的。 + +--- + +### 8. 光线与色彩 + +* 城市阴影; +* 冷灰天光; +* 湿冷反光; +* 人物脸上有轻微侧光; +* 不要强戏剧光; +* 不要暖色柔光; +* 不要男主写真布光。 + +色彩仍然是: + +* 钢铁灰; +* 混凝土灰; +* 深色外套; +* 少量皮肤暗暖。 + +--- + +### 9. 与前后镜关系 + +#### 前一镜 S-13 + +手掌、背包带、掌心的茧,身体细节出现。 + +#### 本镜 S-14 + +人物脸部气质第一次被确认。 + +#### 后一镜 S-15 + +镜头重新拉回长路,“别问终点”,人物再次回到大环境里。 + +这很重要: +**S-14 不能让全片从这里开始变成人物片。** +它只是一次短暂确认。 + +--- + +### 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* middle-aged male traveler around 50; +* half-profile, not portrait pose; +* tired but dignified; +* stubble, slightly messy dark hair; +* dark long coat, backpack strap visible; +* cold urban edge background; +* no glamour lighting; +* no direct hero pose。 + +这不是最终 Prompt,只是执行要点。 + +--- + +### 11. 验收关注点 + +审这一镜看 7 件事: + +1. 年龄是否对? +2. 胡茬和发型是否与人物锚点板一致? +3. 疲惫感有没有? +4. 有没有保持尊严感? +5. 有没有男模写真感? +6. 背包肩带 / 外套是否仍然稳定? +7. 看完这一镜,是否仍相信他就是 S-05 的那个背影? + +--- + +## 镜头 S-21:那一簇火,指引方向 + +### 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 那一簇火 +> 指引方向 + +时间位置约: + +> **1:05 - 1:59** + +这是第一簇火真正变成“方向感”的镜头。 + +--- + +### 2. 本镜情绪功能 + +这一镜不是休息,不是露营。 + +它要表达的是: + +> 火没有让他停下。 +> 火只是让他确认,自己还能继续往前走。 + +这是一枚过渡镜头: +从孤独长路,过渡到夜路中的微弱支撑。 + +--- + +### 3. 核心画面 + +夜色或黄昏后的路边,小火照亮一小段泥路。 + +主角在火光边缘: + +* 没有围着火坐下; +* 没有享受露营; +* 只是停顿、调整背包; +* 侧背对镜头; +* 然后继续向火光照亮的方向走。 + +画面里火要小。 +火不是主角,火是方向。 + +--- + +### 4. 人物与镜头关系 + +#### 人物可见度 + +* 中远景; +* 侧背影; +* 不露脸; +* 可看见外套、背包、靴子; +* 有调整背包的动作。 + +#### 人物占比 + +建议: + +> **1/6 - 1/4** + +因为要看清人物动作,但不能变成篝火人像。 + +--- + +### 5. 人物一致性锚点 + +这一镜验证动态一致性: + +1. **侧背轮廓** + 外套、背包、肩线要和 S-05 保持一致。 + +2. **调整背包动作** + 动作要自然,像走累后重新背稳,不是摆 pose。 + +3. **外套与火光关系** + 火光照亮局部,但不能改变服装主色判断。 + +4. **靴子与泥路** + 脚下仍是路,不是露营地毯或干净草地。 + +5. **行走方向** + 动作必须指向继续前行。 + +--- + +### 6. 场景与视觉母题 + +调用: + +* 夜路; +* 小篝火; +* 旧背包; +* 泥路; +* 火光; +* 继续前行。 + +这是 D 夜路篝火型的前置版本。 +它还不是结尾“远处同行者火光”,只是第一簇火。 + +--- + +### 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 火在画面一侧或下方; +* 人物在火光边缘,不在中心; +* 路从火光旁延伸出去; +* 夜色和道路留出空间; +* 人物动作朝向路,而不是朝向火。 + +构图要让人感觉: + +> 火在身后或身侧,路在前方。 + +--- + +### 8. 光线与色彩 + +* 深蓝灰夜色; +* 小范围暖橙火光; +* 火光照亮泥地、靴子、背包边缘; +* 暗部保留; +* 远处不宜太亮。 + +注意: +火光不能把画面变成温馨露营片。 +它只能给人物一点点温度。 + +--- + +### 9. 与前后镜关系 + +#### 前一镜 S-20 + +火近一点,照亮泥地和石头。 + +#### 本镜 S-21 + +主角调整背包,借火光确认方向,继续上路。 + +#### 后一镜 S-22 + +雨后微光、水洼反光,进入更开阔的继续感。 + +--- + +### 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* small real campfire; +* night roadside, muddy road; +* middle-aged traveler seen from side/back; +* adjusting old backpack; +* continuing forward, not camping; +* dark long coat, worn boots; +* restrained warm firelight; +* no group, no camping advertisement. + +这不是最终 Prompt,只是执行要点。 + +--- + +### 11. 验收关注点 + +审这一镜看 7 件事: + +1. 主角是不是和 S-05 的背影同一个人? +2. 背包、外套、靴子是否稳定? +3. 动作是否自然? +4. 火有没有太大、太温馨? +5. 有没有露营广告感? +6. 人物有没有停留成摆拍? +7. 这镜是否明确表达“火指引方向,而不是火成为终点”? + +--- + +## 镜头 S-25:我擦拭微尘 + +### 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 我擦拭微尘 + +时间位置约: + +> **1:59 - 3:24 主体副歌** + +这是副歌里一个身体动作镜头。 + +--- + +### 2. 本镜情绪功能 + +这一镜不是“擦脸耍帅”。 + +它表达的是: + +> 他不是干净地走过这条路。 +> 他身上有尘土,有疲惫,有走过泥泞后的痕迹。 + +这是一种克制的身体证明。 + +--- + +### 3. 核心画面 + +主角在风里短暂停下,用手擦过: + +* 脸侧; +* 额头; +* 外套肩部; +* 或衣领上的尘土。 + +画面可以只露半侧脸和手。 +手上有粗糙感,外套有尘土,风吹乱头发。 + +不要让动作变得戏剧化。 +它应该很短、很自然,像一个人走累后下意识做的动作。 + +--- + +### 4. 人物与镜头关系 + +#### 人物可见度 + +* 局部; +* 半侧脸; +* 手部动作; +* 不完整正脸; +* 不直视镜头。 + +#### 人物占比 + +建议: + +> **1/3 左右以内** + +这是人物局部动作镜头,可以稍近,但仍要克制。 + +--- + +### 5. 人物一致性锚点 + +这一镜验证的是近距离细节: + +1. **手部质感** + 有中年手的粗糙感,不精致,不干净得像广告手。 + +2. **脸侧轮廓** + 和 S-14、S-33 的脸部气质一致。 + +3. **胡茬** + 明显但自然,不时尚修饰。 + +4. **发型** + 深色、稍乱,被风吹过。 + +5. **外套材质** + 有尘土和使用痕迹,但不破烂。 + +--- + +### 6. 场景与视觉母题 + +调用: + +* 微尘; +* 手; +* 外套; +* 风; +* 中年行者; +* 身体疲惫。 + +这一镜是人物局部和环境痕迹的结合。 +它要让观众感到“这一路真的落在身体上”。 + +--- + +### 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 半侧构图; +* 手部动作在画面前景或中景; +* 脸不要完整正对; +* 背景虚化但能感到旷野或荒路; +* 不要干净影棚感。 + +画面应该像: + +> 镜头经过他身边时,捕捉到一个短暂的下意识动作。 + +--- + +### 8. 光线与色彩 + +* 风沙环境下的冷灰自然光; +* 或黄昏前的低饱和光; +* 皮肤不要过度磨皮; +* 尘土可见但不脏乱; +* 外套仍为深灰 / 暗卡其 / 煤黑。 + +不要: + +* 柔光人像; +* 皮肤精修; +* 明显摆拍; +* 过度煽情眼神。 + +--- + +### 9. 与前后镜关系 + +#### 前一镜 S-24 + +人物在无名旷野里被世界吞没。 + +#### 本镜 S-25 + +镜头拉近到身体痕迹,证明这条路真的磨过他。 + +#### 后一镜 S-26 + +进入“仰望碎裂星海”的铺垫,空间再次打开。 + +这个节奏是: + +> 大远景压小人 → 局部看到尘土 → 再转向星海。 + +--- + +### 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* half-profile middle-aged traveler; +* hand wiping dust from face or coat; +* rough hand, stubble, messy dark hair; +* dark weathered long coat; +* wind and dust; +* natural candid action; +* not a portrait, not glamour. + +这不是最终 Prompt,只是执行要点。 + +--- + +### 11. 验收关注点 + +审这一镜看 7 件事: + +1. 动作是否自然? +2. 手是否有中年粗粝感? +3. 脸侧是否和 S-14 同一个人? +4. 胡茬和发型是否稳定? +5. 有没有擦脸写真感? +6. 尘土是否真实但不过脏? +7. 这个动作是否服务“走过风尘”,而不是服务人物好看? + +--- + +## 镜头 S-33:我咬紧牙关 + +### 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 我咬紧牙关 + +时间位置约: + +> **1:59 - 3:24 主体副歌后段** + +这是全片少数可以接近人物情绪的镜头之一。 + +--- + +### 2. 本镜情绪功能 + +这一镜要表达的是内在用力。 + +不是怒吼。 +不是热血爆发。 +不是“我要逆袭”。 + +而是: + +> 他很累,但没有松。 +> 他咬住了自己,继续走。 + +这是一种中年人的硬撑,不是年轻人的燃爆。 + +--- + +### 3. 核心画面 + +半侧脸短近景。 + +画面中可以看到: + +* 被风吹乱的深色头发; +* 胡茬; +* 嘴角或下颌微微收紧; +* 眼神疲惫但有火; +* 深色外套领口; +* 背景是模糊的旷野、夜路或风沙。 + +注意: +“咬紧牙关”不一定要真的露牙。 +更建议用下颌、嘴角、眼神和呼吸状态表达。 + +--- + +### 4. 人物与镜头关系 + +#### 人物可见度 + +* 半侧脸; +* 短近景; +* 不正面直视; +* 不做表演式表情。 + +#### 人物占比 + +这一镜可以是人物占比最高的镜头之一: + +> **1/3 - 1/2** + +但必须短、克制,不能让它变成人物情绪写真。 + +--- + +### 5. 人物一致性锚点 + +这一镜是人物脸部一致性最重要的测试之一。 + +必须验证: + +1. **是否还是 S-14 的同一个人** +2. **年龄感是否稳定** +3. **胡茬是否一致** +4. **发型是否一致** +5. **眼神是否仍是“疲惫但有火”** +6. **有没有变成硬汉海报** +7. **有没有年轻化或帅化** + +这一镜最容易跑偏,因为它最接近人物脸。 + +--- + +### 6. 场景与视觉母题 + +调用: + +* 风沙; +* 中年行者; +* 胡茬; +* 咬紧牙关; +* 内在用力; +* 继续。 + +这不是风景镜头,但它必须仍属于风景叙事 MV。 +背景不能完全消失成影棚人像。 + +--- + +### 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 半侧脸在画面一侧; +* 留出风、夜色或荒路背景; +* 头发和外套被风轻微影响; +* 不要正脸居中; +* 不要强眼神直视镜头。 + +镜头感觉应该是: + +> 人在路上经过一瞬间,镜头捕捉到他把自己撑住的表情。 + +--- + +### 8. 光线与色彩 + +可选两种光线: + +#### 方案 A:冷风沙光 + +* 灰蓝自然光; +* 脸部暗部保留; +* 风尘感明显; +* 适合“咬紧牙关”的硬度。 + +#### 方案 B:夜路微火边缘光 + +* 一点点暖火光扫过脸侧; +* 背景仍是深蓝黑; +* 情绪更温暖,但要避免煽情。 + +两种都可以,关键是不要: + +* 人像棚拍光; +* 英雄硬光; +* 精致电影男主光; +* 过度磨皮。 + +--- + +### 9. 与前后镜关系 + +#### 前一镜 S-32 + +远处脚印或微弱火点暗示同行者。 + +#### 本镜 S-33 + +镜头靠近主角,看到他内在用力的一瞬间。 + +#### 后一镜 S-34 + +镜头落到靴跟和泥路,人物情绪立刻转回身体行动。 + +这个节奏很好: + +> 远方有人 → 他咬住自己 → 脚继续走。 + +--- + +### 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* half-profile close shot; +* middle-aged male traveler around 50; +* tired but determined; +* stubble, slightly messy dark hair; +* jaw tightened, restrained emotion; +* dark long coat collar visible; +* windy road background; +* no heroic poster pose; +* no glamour portrait. + +这不是最终 Prompt,只是执行要点。 + +--- + +### 11. 验收关注点 + +审这一镜看 8 件事: + +1. 是否和 S-14 是同一个人? +2. 是否保留 50 岁左右的中年感? +3. 胡茬、发型、外套是否稳定? +4. 有没有“疲惫但有火”? +5. 有没有表演过度? +6. 有没有英雄海报感? +7. 有没有精致男模化? +8. 看完这一镜,是否更相信他会继续走,而不是觉得他在摆拍? + +--- + +## 6. 这 5 镜的联动关系 + +这 5 个镜头组合起来,正好是一套人物一致性验证链: + +### S-05:先认背影 + +观众先认识他的背包、外套和肩背轮廓。 + +### S-21:再认动态 + +观众看到同一个人如何调整背包、继续走。 + +### S-14:确认脸部气质 + +观众第一次知道这个行者是怎样的中年人。 + +### S-25:确认身体痕迹 + +观众看到风尘、手、微尘,确认这条路落在身体上。 + +### S-33:确认精神状态 + +观众看到他最接近崩但没崩的一瞬间。 + +它们共同建立: + +> 他不是一个漂亮角色。 +> 他是一个被路磨过、但还在走的人。 + +--- + +## 7. 本轮测试的核心验收标准 + +这 5 镜测试完,不要只看单张美不美。 +重点看 5 个统一性问题: + +### 1. 背影和脸是不是同一个人? + +S-05 的背影,能不能自然连接到 S-14 和 S-33 的脸。 + +### 2. 服装有没有稳定? + +长外套、旧背包、靴子不能每镜都变。 + +### 3. 发型和胡茬有没有稳定? + +尤其 S-14、S-25、S-33 三个靠近脸的镜头。 + +### 4. 气质有没有稳定? + +始终是: + +> 沧桑摇滚,温暖疲惫,孤独但坚定。 + +不能一会儿流浪汉,一会儿男模,一会儿硬汉主角。 + +### 5. 人物有没有抢走 MV 的风景叙事基调? + +即便 S-33 是近一点的情绪镜头,也不能让全片变成人物写真。 + +--- + +## 8. 五镜最终判断标准 + +如果这 5 个镜头成立,人物系统基本就过了第一关。 + +因为它们覆盖了: + +* 背影; +* 侧背; +* 动作; +* 半侧脸; +* 手部; +* 情绪近点; +* 道具; +* 外套; +* 发型; +* 胡茬; +* 年龄; +* 气质。 + +也就是说,后面再做远景、城市、火光、水洼,就不容易“每张像不同人”。 + +本轮核心目标可以概括为一句: + +> **让观众不靠正脸,也能相信:这一路上走着的,始终是同一个人。** diff --git a/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-map.json b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-map.json new file mode 100644 index 0000000..ec5ad4a --- /dev/null +++ b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-map.json @@ -0,0 +1,516 @@ +{ + "source": "C:\\Users\\wangq\\Documents\\Codex\\video-workbench\\projects\\2026-06-23-在路上\\intake\\阶段五-人物一致性单镜.md", + "heading_count": 102, + "headings": [ + { + "line": 1, + "level": 1, + "text": "《在路上》人物一致性验证单镜骨架 V1" + }, + { + "line": 17, + "level": 1, + "text": "0. 这 5 镜的人物一致性验证分工" + }, + { + "line": 35, + "level": 1, + "text": "镜头 S-05:斑驳行囊" + }, + { + "line": 37, + "level": 2, + "text": "1. 对应歌词 / 音乐段落" + }, + { + "line": 51, + "level": 2, + "text": "2. 本镜情绪功能" + }, + { + "line": 63, + "level": 2, + "text": "3. 核心画面" + }, + { + "line": 81, + "level": 2, + "text": "4. 人物与镜头关系" + }, + { + "line": 83, + "level": 3, + "text": "人物可见度" + }, + { + "line": 90, + "level": 3, + "text": "人物占比" + }, + { + "line": 100, + "level": 2, + "text": "5. 人物一致性锚点" + }, + { + "line": 121, + "level": 2, + "text": "6. 场景与视觉母题" + }, + { + "line": 136, + "level": 2, + "text": "7. 构图方向" + }, + { + "line": 152, + "level": 2, + "text": "8. 光线与色彩" + }, + { + "line": 166, + "level": 2, + "text": "9. 与前后镜关系" + }, + { + "line": 168, + "level": 3, + "text": "前一镜 S-04" + }, + { + "line": 172, + "level": 3, + "text": "本镜 S-05" + }, + { + "line": 176, + "level": 3, + "text": "后一镜 S-06" + }, + { + "line": 182, + "level": 2, + "text": "10. 执行层要点" + }, + { + "line": 199, + "level": 2, + "text": "11. 验收关注点" + }, + { + "line": 212, + "level": 1, + "text": "镜头 S-14:疲惫的人 / 你醒了吗" + }, + { + "line": 214, + "level": 2, + "text": "1. 对应歌词 / 音乐段落" + }, + { + "line": 229, + "level": 2, + "text": "2. 本镜情绪功能" + }, + { + "line": 242, + "level": 2, + "text": "3. 核心画面" + }, + { + "line": 260, + "level": 2, + "text": "4. 人物与镜头关系" + }, + { + "line": 262, + "level": 3, + "text": "人物可见度" + }, + { + "line": 269, + "level": 3, + "text": "人物占比" + }, + { + "line": 279, + "level": 2, + "text": "5. 人物一致性锚点" + }, + { + "line": 300, + "level": 2, + "text": "6. 场景与视觉母题" + }, + { + "line": 316, + "level": 2, + "text": "7. 构图方向" + }, + { + "line": 333, + "level": 2, + "text": "8. 光线与色彩" + }, + { + "line": 352, + "level": 2, + "text": "9. 与前后镜关系" + }, + { + "line": 354, + "level": 3, + "text": "前一镜 S-13" + }, + { + "line": 358, + "level": 3, + "text": "本镜 S-14" + }, + { + "line": 362, + "level": 3, + "text": "后一镜 S-15" + }, + { + "line": 372, + "level": 2, + "text": "10. 执行层要点" + }, + { + "line": 389, + "level": 2, + "text": "11. 验收关注点" + }, + { + "line": 403, + "level": 1, + "text": "镜头 S-21:那一簇火,指引方向" + }, + { + "line": 405, + "level": 2, + "text": "1. 对应歌词 / 音乐段落" + }, + { + "line": 420, + "level": 2, + "text": "2. 本镜情绪功能" + }, + { + "line": 434, + "level": 2, + "text": "3. 核心画面" + }, + { + "line": 451, + "level": 2, + "text": "4. 人物与镜头关系" + }, + { + "line": 453, + "level": 3, + "text": "人物可见度" + }, + { + "line": 461, + "level": 3, + "text": "人物占比" + }, + { + "line": 471, + "level": 2, + "text": "5. 人物一致性锚点" + }, + { + "line": 492, + "level": 2, + "text": "6. 场景与视觉母题" + }, + { + "line": 508, + "level": 2, + "text": "7. 构图方向" + }, + { + "line": 525, + "level": 2, + "text": "8. 光线与色彩" + }, + { + "line": 539, + "level": 2, + "text": "9. 与前后镜关系" + }, + { + "line": 541, + "level": 3, + "text": "前一镜 S-20" + }, + { + "line": 545, + "level": 3, + "text": "本镜 S-21" + }, + { + "line": 549, + "level": 3, + "text": "后一镜 S-22" + }, + { + "line": 555, + "level": 2, + "text": "10. 执行层要点" + }, + { + "line": 572, + "level": 2, + "text": "11. 验收关注点" + }, + { + "line": 586, + "level": 1, + "text": "镜头 S-25:我擦拭微尘" + }, + { + "line": 588, + "level": 2, + "text": "1. 对应歌词 / 音乐段落" + }, + { + "line": 602, + "level": 2, + "text": "2. 本镜情绪功能" + }, + { + "line": 615, + "level": 2, + "text": "3. 核心画面" + }, + { + "line": 632, + "level": 2, + "text": "4. 人物与镜头关系" + }, + { + "line": 634, + "level": 3, + "text": "人物可见度" + }, + { + "line": 642, + "level": 3, + "text": "人物占比" + }, + { + "line": 652, + "level": 2, + "text": "5. 人物一致性锚点" + }, + { + "line": 673, + "level": 2, + "text": "6. 场景与视觉母题" + }, + { + "line": 689, + "level": 2, + "text": "7. 构图方向" + }, + { + "line": 706, + "level": 2, + "text": "8. 光线与色彩" + }, + { + "line": 723, + "level": 2, + "text": "9. 与前后镜关系" + }, + { + "line": 725, + "level": 3, + "text": "前一镜 S-24" + }, + { + "line": 729, + "level": 3, + "text": "本镜 S-25" + }, + { + "line": 733, + "level": 3, + "text": "后一镜 S-26" + }, + { + "line": 743, + "level": 2, + "text": "10. 执行层要点" + }, + { + "line": 759, + "level": 2, + "text": "11. 验收关注点" + }, + { + "line": 773, + "level": 1, + "text": "镜头 S-33:我咬紧牙关" + }, + { + "line": 775, + "level": 2, + "text": "1. 对应歌词 / 音乐段落" + }, + { + "line": 789, + "level": 2, + "text": "2. 本镜情绪功能" + }, + { + "line": 806, + "level": 2, + "text": "3. 核心画面" + }, + { + "line": 825, + "level": 2, + "text": "4. 人物与镜头关系" + }, + { + "line": 827, + "level": 3, + "text": "人物可见度" + }, + { + "line": 834, + "level": 3, + "text": "人物占比" + }, + { + "line": 844, + "level": 2, + "text": "5. 人物一致性锚点" + }, + { + "line": 862, + "level": 2, + "text": "6. 场景与视觉母题" + }, + { + "line": 878, + "level": 2, + "text": "7. 构图方向" + }, + { + "line": 895, + "level": 2, + "text": "8. 光线与色彩" + }, + { + "line": 899, + "level": 3, + "text": "方案 A:冷风沙光" + }, + { + "line": 906, + "level": 3, + "text": "方案 B:夜路微火边缘光" + }, + { + "line": 921, + "level": 2, + "text": "9. 与前后镜关系" + }, + { + "line": 923, + "level": 3, + "text": "前一镜 S-32" + }, + { + "line": 927, + "level": 3, + "text": "本镜 S-33" + }, + { + "line": 931, + "level": 3, + "text": "后一镜 S-34" + }, + { + "line": 941, + "level": 2, + "text": "10. 执行层要点" + }, + { + "line": 959, + "level": 2, + "text": "11. 验收关注点" + }, + { + "line": 974, + "level": 1, + "text": "6. 这 5 镜的联动关系" + }, + { + "line": 978, + "level": 2, + "text": "S-05:先认背影" + }, + { + "line": 982, + "level": 2, + "text": "S-21:再认动态" + }, + { + "line": 986, + "level": 2, + "text": "S-14:确认脸部气质" + }, + { + "line": 990, + "level": 2, + "text": "S-25:确认身体痕迹" + }, + { + "line": 994, + "level": 2, + "text": "S-33:确认精神状态" + }, + { + "line": 1005, + "level": 1, + "text": "7. 本轮测试的核心验收标准" + }, + { + "line": 1010, + "level": 2, + "text": "1. 背影和脸是不是同一个人?" + }, + { + "line": 1014, + "level": 2, + "text": "2. 服装有没有稳定?" + }, + { + "line": 1018, + "level": 2, + "text": "3. 发型和胡茬有没有稳定?" + }, + { + "line": 1022, + "level": 2, + "text": "4. 气质有没有稳定?" + }, + { + "line": 1030, + "level": 2, + "text": "5. 人物有没有抢走 MV 的风景叙事基调?" + }, + { + "line": 1036, + "level": 1, + "text": "8. 五镜最终判断标准" + } + ] +} diff --git a/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-plan.json b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-plan.json new file mode 100644 index 0000000..1ed0112 --- /dev/null +++ b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-plan.json @@ -0,0 +1,618 @@ +{ + "mode": "artifact", + "source": "C:\\Users\\wangq\\Documents\\Codex\\video-workbench\\projects\\2026-06-23-在路上\\intake\\阶段五-人物一致性单镜.md", + "output": "C:\\Users\\wangq\\Documents\\Codex\\video-workbench\\projects\\2026-06-23-在路上\\intake-repair\\fix-title-2026-06-24-character-consistency\\阶段五-人物一致性单镜.fixed.md", + "report": "C:\\Users\\wangq\\Documents\\Codex\\video-workbench\\projects\\2026-06-23-在路上\\intake-repair\\fix-title-2026-06-24-character-consistency\\阶段五-人物一致性单镜.heading-report.md", + "document_title_line": 1, + "document_title": "《在路上》人物一致性验证单镜骨架 V1", + "semantic_summary": "Preserve the first visible H1 as the only document title. Treat all later H1 headings as major sections under that title and recursively nest their children one level deeper.", + "ambiguities": [], + "edits": [ + { + "line": 17, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 35, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 37, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 51, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 63, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 81, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 83, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 90, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 100, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 121, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 136, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 152, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 166, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 168, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 172, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 176, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 182, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 199, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 212, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 214, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 229, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 242, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 260, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 262, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 269, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 279, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 300, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 316, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 333, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 352, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 354, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 358, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 362, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 372, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 389, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 403, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 405, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 420, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 434, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 451, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 453, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 461, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 471, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 492, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 508, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 525, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 539, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 541, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 545, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 549, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 555, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 572, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 586, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 588, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 602, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 615, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 632, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 634, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 642, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 652, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 673, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 689, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 706, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 723, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 725, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 729, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 733, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 743, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 759, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 773, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 775, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 789, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 806, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 825, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 827, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 834, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 844, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 862, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 878, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 895, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 899, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 906, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 921, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 923, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 927, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 931, + "from_level": 3, + "to_level": 4, + "reason": "This detail heading belongs under the preceding subsection after recursive nesting." + }, + { + "line": 941, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 959, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 974, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 978, + "from_level": 2, + "to_level": 3, + "reason": "This is a child item inside the summary/linkage section after its parent is nested under the document title." + }, + { + "line": 982, + "from_level": 2, + "to_level": 3, + "reason": "This is a child item inside the summary/linkage section after its parent is nested under the document title." + }, + { + "line": 986, + "from_level": 2, + "to_level": 3, + "reason": "This is a child item inside the summary/linkage section after its parent is nested under the document title." + }, + { + "line": 990, + "from_level": 2, + "to_level": 3, + "reason": "This is a child item inside the summary/linkage section after its parent is nested under the document title." + }, + { + "line": 994, + "from_level": 2, + "to_level": 3, + "reason": "This is a child item inside the summary/linkage section after its parent is nested under the document title." + }, + { + "line": 1005, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + }, + { + "line": 1010, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 1014, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 1018, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 1022, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 1030, + "from_level": 2, + "to_level": 3, + "reason": "This subsection belongs inside the current major section after that section is nested under the document title." + }, + { + "line": 1036, + "from_level": 1, + "to_level": 2, + "reason": "This is a major document section under the single document title, not a separate artifact title." + } + ] +} diff --git a/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-report.md b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-report.md new file mode 100644 index 0000000..416355f --- /dev/null +++ b/projects/2026-06-23-在路上/intake-repair/fix-title-2026-06-24-character-consistency/阶段五-人物一致性单镜.heading-report.md @@ -0,0 +1,121 @@ +# Semantic Heading Repair Report + +- Source: `C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake\阶段五-人物一致性单镜.md` +- Output: `C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake-repair\fix-title-2026-06-24-character-consistency\阶段五-人物一致性单镜.fixed.md` +- Plan: `C:\Users\wangq\Documents\Codex\video-workbench\projects\2026-06-23-在路上\intake-repair\fix-title-2026-06-24-character-consistency\阶段五-人物一致性单镜.heading-plan.json` +- Mode: `artifact` +- Changed headings: 101 + +| Line | From | To | Heading | Reason | +| --- | --- | --- | --- | --- | +| 17 | 1 | 2 | 0. 这 5 镜的人物一致性验证分工 | This is a major document section under the single document title, not a separate artifact title. | +| 35 | 1 | 2 | 镜头 S-05:斑驳行囊 | This is a major document section under the single document title, not a separate artifact title. | +| 37 | 2 | 3 | 1. 对应歌词 / 音乐段落 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 51 | 2 | 3 | 2. 本镜情绪功能 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 63 | 2 | 3 | 3. 核心画面 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 81 | 2 | 3 | 4. 人物与镜头关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 83 | 3 | 4 | 人物可见度 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 90 | 3 | 4 | 人物占比 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 100 | 2 | 3 | 5. 人物一致性锚点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 121 | 2 | 3 | 6. 场景与视觉母题 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 136 | 2 | 3 | 7. 构图方向 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 152 | 2 | 3 | 8. 光线与色彩 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 166 | 2 | 3 | 9. 与前后镜关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 168 | 3 | 4 | 前一镜 S-04 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 172 | 3 | 4 | 本镜 S-05 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 176 | 3 | 4 | 后一镜 S-06 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 182 | 2 | 3 | 10. 执行层要点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 199 | 2 | 3 | 11. 验收关注点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 212 | 1 | 2 | 镜头 S-14:疲惫的人 / 你醒了吗 | This is a major document section under the single document title, not a separate artifact title. | +| 214 | 2 | 3 | 1. 对应歌词 / 音乐段落 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 229 | 2 | 3 | 2. 本镜情绪功能 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 242 | 2 | 3 | 3. 核心画面 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 260 | 2 | 3 | 4. 人物与镜头关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 262 | 3 | 4 | 人物可见度 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 269 | 3 | 4 | 人物占比 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 279 | 2 | 3 | 5. 人物一致性锚点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 300 | 2 | 3 | 6. 场景与视觉母题 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 316 | 2 | 3 | 7. 构图方向 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 333 | 2 | 3 | 8. 光线与色彩 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 352 | 2 | 3 | 9. 与前后镜关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 354 | 3 | 4 | 前一镜 S-13 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 358 | 3 | 4 | 本镜 S-14 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 362 | 3 | 4 | 后一镜 S-15 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 372 | 2 | 3 | 10. 执行层要点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 389 | 2 | 3 | 11. 验收关注点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 403 | 1 | 2 | 镜头 S-21:那一簇火,指引方向 | This is a major document section under the single document title, not a separate artifact title. | +| 405 | 2 | 3 | 1. 对应歌词 / 音乐段落 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 420 | 2 | 3 | 2. 本镜情绪功能 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 434 | 2 | 3 | 3. 核心画面 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 451 | 2 | 3 | 4. 人物与镜头关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 453 | 3 | 4 | 人物可见度 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 461 | 3 | 4 | 人物占比 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 471 | 2 | 3 | 5. 人物一致性锚点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 492 | 2 | 3 | 6. 场景与视觉母题 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 508 | 2 | 3 | 7. 构图方向 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 525 | 2 | 3 | 8. 光线与色彩 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 539 | 2 | 3 | 9. 与前后镜关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 541 | 3 | 4 | 前一镜 S-20 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 545 | 3 | 4 | 本镜 S-21 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 549 | 3 | 4 | 后一镜 S-22 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 555 | 2 | 3 | 10. 执行层要点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 572 | 2 | 3 | 11. 验收关注点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 586 | 1 | 2 | 镜头 S-25:我擦拭微尘 | This is a major document section under the single document title, not a separate artifact title. | +| 588 | 2 | 3 | 1. 对应歌词 / 音乐段落 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 602 | 2 | 3 | 2. 本镜情绪功能 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 615 | 2 | 3 | 3. 核心画面 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 632 | 2 | 3 | 4. 人物与镜头关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 634 | 3 | 4 | 人物可见度 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 642 | 3 | 4 | 人物占比 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 652 | 2 | 3 | 5. 人物一致性锚点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 673 | 2 | 3 | 6. 场景与视觉母题 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 689 | 2 | 3 | 7. 构图方向 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 706 | 2 | 3 | 8. 光线与色彩 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 723 | 2 | 3 | 9. 与前后镜关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 725 | 3 | 4 | 前一镜 S-24 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 729 | 3 | 4 | 本镜 S-25 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 733 | 3 | 4 | 后一镜 S-26 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 743 | 2 | 3 | 10. 执行层要点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 759 | 2 | 3 | 11. 验收关注点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 773 | 1 | 2 | 镜头 S-33:我咬紧牙关 | This is a major document section under the single document title, not a separate artifact title. | +| 775 | 2 | 3 | 1. 对应歌词 / 音乐段落 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 789 | 2 | 3 | 2. 本镜情绪功能 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 806 | 2 | 3 | 3. 核心画面 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 825 | 2 | 3 | 4. 人物与镜头关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 827 | 3 | 4 | 人物可见度 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 834 | 3 | 4 | 人物占比 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 844 | 2 | 3 | 5. 人物一致性锚点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 862 | 2 | 3 | 6. 场景与视觉母题 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 878 | 2 | 3 | 7. 构图方向 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 895 | 2 | 3 | 8. 光线与色彩 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 899 | 3 | 4 | 方案 A:冷风沙光 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 906 | 3 | 4 | 方案 B:夜路微火边缘光 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 921 | 2 | 3 | 9. 与前后镜关系 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 923 | 3 | 4 | 前一镜 S-32 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 927 | 3 | 4 | 本镜 S-33 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 931 | 3 | 4 | 后一镜 S-34 | This detail heading belongs under the preceding subsection after recursive nesting. | +| 941 | 2 | 3 | 10. 执行层要点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 959 | 2 | 3 | 11. 验收关注点 | This subsection belongs inside the current major section after that section is nested under the document title. | +| 974 | 1 | 2 | 6. 这 5 镜的联动关系 | This is a major document section under the single document title, not a separate artifact title. | +| 978 | 2 | 3 | S-05:先认背影 | This is a child item inside the summary/linkage section after its parent is nested under the document title. | +| 982 | 2 | 3 | S-21:再认动态 | This is a child item inside the summary/linkage section after its parent is nested under the document title. | +| 986 | 2 | 3 | S-14:确认脸部气质 | This is a child item inside the summary/linkage section after its parent is nested under the document title. | +| 990 | 2 | 3 | S-25:确认身体痕迹 | This is a child item inside the summary/linkage section after its parent is nested under the document title. | +| 994 | 2 | 3 | S-33:确认精神状态 | This is a child item inside the summary/linkage section after its parent is nested under the document title. | +| 1005 | 1 | 2 | 7. 本轮测试的核心验收标准 | This is a major document section under the single document title, not a separate artifact title. | +| 1010 | 2 | 3 | 1. 背影和脸是不是同一个人? | This subsection belongs inside the current major section after that section is nested under the document title. | +| 1014 | 2 | 3 | 2. 服装有没有稳定? | This subsection belongs inside the current major section after that section is nested under the document title. | +| 1018 | 2 | 3 | 3. 发型和胡茬有没有稳定? | This subsection belongs inside the current major section after that section is nested under the document title. | +| 1022 | 2 | 3 | 4. 气质有没有稳定? | This subsection belongs inside the current major section after that section is nested under the document title. | +| 1030 | 2 | 3 | 5. 人物有没有抢走 MV 的风景叙事基调? | This subsection belongs inside the current major section after that section is nested under the document title. | +| 1036 | 1 | 2 | 8. 五镜最终判断标准 | This is a major document section under the single document title, not a separate artifact title. | + +## Semantic Decision Summary + +- Preserved line 1 as the single document title. +- Treated the later shot blocks and numbered summary sections as peer major sections under that title. +- Recursively nested subsection headings under their repaired parents. + +## Ambiguities + +- None detected. No appended second artifact was identified. diff --git a/projects/2026-06-23-在路上/intake/阶段五-人物一致性单镜.md b/projects/2026-06-23-在路上/intake/阶段五-人物一致性单镜.md new file mode 100644 index 0000000..5543c5d --- /dev/null +++ b/projects/2026-06-23-在路上/intake/阶段五-人物一致性单镜.md @@ -0,0 +1,1059 @@ +# 《在路上》人物一致性验证单镜骨架 V1 + +本次展开 5 个镜头: + +* **S-05:斑驳行囊** +* **S-14:疲惫的人 / 你醒了吗** +* **S-21:那一簇火,指引方向** +* **S-25:我擦拭微尘** +* **S-33:我咬紧牙关** + +这 5 个镜头的作用,不只是服务歌词,而是共同验证一个问题: + +> 在背影、侧背、半侧脸、手部动作、情绪近点里,主角是不是仍然像同一个“路上行者”。 + +--- + +# 0. 这 5 镜的人物一致性验证分工 + +| 镜头 | 验证重点 | 人物可见度 | 主要锚点 | +| ---- | -------- | --------- | ------------ | +| S-05 | 背影与行囊一致性 | 背影局部 / 中近 | 背包、外套下摆、肩背轮廓 | +| S-21 | 侧背动态一致性 | 中远景 / 侧背影 | 调整背包、火光、继续前行 | +| S-14 | 半侧脸气质一致性 | 半身偏侧 | 年龄、胡茬、疲惫但不塌 | +| S-25 | 手部动作与微尘感 | 局部 / 半侧 | 手、脸侧、外套、风尘感 | +| S-33 | 情绪近点一致性 | 半侧脸短近景 | 牙关、胡茬、眼神有火 | + +测试顺序建议: + +> **S-05 → S-21 → S-14 → S-25 → S-33** + +先从背影和服装稳定性开始,再逐步靠近人物脸部,避免一上来陷入“像不像证件照”的判断。 + +--- + +# 镜头 S-05:斑驳行囊 + +## 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 斑驳行囊 + +时间位置约: + +> **0:16 - 0:32** + +这是早段人物锚点第一次被观众看清的镜头。 + +--- + +## 2. 本镜情绪功能 + +这个镜头不负责表现人物脸。 +它负责告诉观众: + +> 这个人在路上已经走了很久。 +> 不是刚出发,不是摆拍,而是真的背着一段时间和重量在走。 + +它也是后续所有背影、远景小人物镜头的服装和道具基准。 + +--- + +## 3. 核心画面 + +清晨冷光里,主角背对镜头行走。 + +画面重点不是脸,而是: + +* 旧背包; +* 深灰 / 暗卡其长外套; +* 外套下摆被风轻轻掀起; +* 背包肩带压在肩上; +* 靴子偶尔进入画面下缘; +* 泥泞路面和风沙背景轻微可见。 + +人物可以是中近的背影局部,也可以是中远景背影。 +但不要拍成“背影写真”。 + +--- + +## 4. 人物与镜头关系 + +### 人物可见度 + +* 背影; +* 不露脸; +* 重点在肩背、背包、外套轮廓; +* 靴子可以部分出现。 + +### 人物占比 + +建议人物占画面: + +> **1/4 左右以内** + +因为这镜是道具锚点,可以比普通远景稍近,但仍不能变成人物写真。 + +--- + +## 5. 人物一致性锚点 + +这一镜必须锁定 5 个稳定元素: + +1. **背包形状** + 旧、实用、体量适中,不像潮流机能包。 + +2. **外套长度** + 长外套下摆要形成背影识别特征。 + +3. **肩背姿态** + 有重量感,肩膀略沉,但不能塌。 + +4. **发型后轮廓** + 深色、稍乱,中年感,不要潮男造型。 + +5. **整体气质** + 风尘、疲惫、继续走,不是旅行博主。 + +--- + +## 6. 场景与视觉母题 + +调用: + +* 清晨薄霜; +* 斑驳行囊; +* 旧背包; +* 长外套; +* 泥路; +* 风。 + +这是人物锚点第一次明确出现,应该和后续人物锚点板保持高度一致。 + +--- + +## 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 背影偏画面一侧; +* 路和环境仍要占一定比例; +* 背包和外套处于视觉中心; +* 不要正面,不要回头。 + +画面应该像: + +> 镜头从后方跟了一小段路,观众第一次注意到这个人背上的重量。 + +--- + +## 8. 光线与色彩 + +* 清晨冷光; +* 低饱和; +* 灰白薄霜; +* 土褐泥路; +* 外套深灰 / 暗卡其; +* 背包旧而暗。 + +不要暖黄旅行片光线。 +不要把背包拍得太新、太干净、太户外广告。 + +--- + +## 9. 与前后镜关系 + +### 前一镜 S-04 + +清晨薄霜,世界先出现。 + +### 本镜 S-05 + +人物锚点出现,但只给背影和行囊。 + +### 后一镜 S-06 + +主角站在岔路前,进入“你要去哪”的方向问题。 + +--- + +## 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* 背影; +* 旧背包; +* 长外套; +* 清晨泥路; +* 行走中; +* 不看镜头; +* 不要肖像感; +* 不要旅行广告感。 + +这不是最终 Prompt,只是执行要点。 + +--- + +## 11. 验收关注点 + +审这一镜时看 6 件事: + +1. 背包是否稳定、可复用? +2. 外套下摆是否形成识别轮廓? +3. 背影是否像“同一个行者”? +4. 有没有旅行写真感? +5. 人物有没有太精致? +6. 这张图缩小后,还能不能认出是这个角色? + +--- + +# 镜头 S-14:疲惫的人 / 你醒了吗 + +## 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 疲惫的人 +> 你醒了吗 + +时间位置约: + +> **0:32 - 1:05 段落末尾** + +这是城市压迫段里的一个人物确认镜头。 + +--- + +## 2. 本镜情绪功能 + +这一镜是全片早段少数可以接近人物的镜头。 +它要完成的是: + +> 让观众第一次确认:这个行者是一个 50 岁左右的中年人,疲惫,但精神没有塌。 + +它不是人物写真。 +不是“拍一个沧桑男人很帅”。 +它是让观众在城市阴影里看见一个被现实磨过的人。 + +--- + +## 3. 核心画面 + +城市边缘阴影里,主角短暂停顿。 + +画面里可以有: + +* 半侧脸; +* 胡茬; +* 被风吹乱的深色头发; +* 深色长外套领口; +* 旧背包肩带; +* 背后是冷硬混凝土、钢铁结构或湿冷墙面。 + +人物不需要直视镜头。 +最好是看向画面外侧或低头后微微抬眼。 + +--- + +## 4. 人物与镜头关系 + +### 人物可见度 + +* 半身偏侧; +* 半侧脸; +* 不完全正脸; +* 不直视镜头或只短暂擦过镜头方向。 + +### 人物占比 + +可以比多数镜头稍大: + +> **1/3 左右以内** + +但仍要保留城市阴影和环境压迫感。 + +--- + +## 5. 人物一致性锚点 + +这一镜验证的是脸部和气质: + +1. **年龄感** + 50 岁左右,不要年轻化。 + +2. **胡茬** + 明显,但不能修成潮流胡型。 + +3. **头发** + 深色,稍乱,有风尘感,不要精致发型。 + +4. **眼神** + 疲惫,但不能空掉;有一点火。 + +5. **气质边界** + 不能像流浪汉,不能像时尚男模,不能像电影硬汉摆拍。 + +--- + +## 6. 场景与视觉母题 + +调用: + +* 城市边缘; +* 钢铁峡谷后的阴影; +* 疲惫的人; +* 中年行者; +* 冷灰色光线。 + +这一镜是 S-10 钢铁峡谷之后的情绪落点: + +> 城市把人压小之后,观众短暂看见这个人的脸。 + +--- + +## 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 人物偏侧,不居中; +* 背后有冷硬城市结构; +* 半侧脸处于暗部和冷光之间; +* 镜头不要太近; +* 不要海报式人像构图。 + +这一镜的关键不是“好看”,而是: + +> 疲惫是真的,尊严也是真的。 + +--- + +## 8. 光线与色彩 + +* 城市阴影; +* 冷灰天光; +* 湿冷反光; +* 人物脸上有轻微侧光; +* 不要强戏剧光; +* 不要暖色柔光; +* 不要男主写真布光。 + +色彩仍然是: + +* 钢铁灰; +* 混凝土灰; +* 深色外套; +* 少量皮肤暗暖。 + +--- + +## 9. 与前后镜关系 + +### 前一镜 S-13 + +手掌、背包带、掌心的茧,身体细节出现。 + +### 本镜 S-14 + +人物脸部气质第一次被确认。 + +### 后一镜 S-15 + +镜头重新拉回长路,“别问终点”,人物再次回到大环境里。 + +这很重要: +**S-14 不能让全片从这里开始变成人物片。** +它只是一次短暂确认。 + +--- + +## 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* middle-aged male traveler around 50; +* half-profile, not portrait pose; +* tired but dignified; +* stubble, slightly messy dark hair; +* dark long coat, backpack strap visible; +* cold urban edge background; +* no glamour lighting; +* no direct hero pose。 + +这不是最终 Prompt,只是执行要点。 + +--- + +## 11. 验收关注点 + +审这一镜看 7 件事: + +1. 年龄是否对? +2. 胡茬和发型是否与人物锚点板一致? +3. 疲惫感有没有? +4. 有没有保持尊严感? +5. 有没有男模写真感? +6. 背包肩带 / 外套是否仍然稳定? +7. 看完这一镜,是否仍相信他就是 S-05 的那个背影? + +--- + +# 镜头 S-21:那一簇火,指引方向 + +## 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 那一簇火 +> 指引方向 + +时间位置约: + +> **1:05 - 1:59** + +这是第一簇火真正变成“方向感”的镜头。 + +--- + +## 2. 本镜情绪功能 + +这一镜不是休息,不是露营。 + +它要表达的是: + +> 火没有让他停下。 +> 火只是让他确认,自己还能继续往前走。 + +这是一枚过渡镜头: +从孤独长路,过渡到夜路中的微弱支撑。 + +--- + +## 3. 核心画面 + +夜色或黄昏后的路边,小火照亮一小段泥路。 + +主角在火光边缘: + +* 没有围着火坐下; +* 没有享受露营; +* 只是停顿、调整背包; +* 侧背对镜头; +* 然后继续向火光照亮的方向走。 + +画面里火要小。 +火不是主角,火是方向。 + +--- + +## 4. 人物与镜头关系 + +### 人物可见度 + +* 中远景; +* 侧背影; +* 不露脸; +* 可看见外套、背包、靴子; +* 有调整背包的动作。 + +### 人物占比 + +建议: + +> **1/6 - 1/4** + +因为要看清人物动作,但不能变成篝火人像。 + +--- + +## 5. 人物一致性锚点 + +这一镜验证动态一致性: + +1. **侧背轮廓** + 外套、背包、肩线要和 S-05 保持一致。 + +2. **调整背包动作** + 动作要自然,像走累后重新背稳,不是摆 pose。 + +3. **外套与火光关系** + 火光照亮局部,但不能改变服装主色判断。 + +4. **靴子与泥路** + 脚下仍是路,不是露营地毯或干净草地。 + +5. **行走方向** + 动作必须指向继续前行。 + +--- + +## 6. 场景与视觉母题 + +调用: + +* 夜路; +* 小篝火; +* 旧背包; +* 泥路; +* 火光; +* 继续前行。 + +这是 D 夜路篝火型的前置版本。 +它还不是结尾“远处同行者火光”,只是第一簇火。 + +--- + +## 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 火在画面一侧或下方; +* 人物在火光边缘,不在中心; +* 路从火光旁延伸出去; +* 夜色和道路留出空间; +* 人物动作朝向路,而不是朝向火。 + +构图要让人感觉: + +> 火在身后或身侧,路在前方。 + +--- + +## 8. 光线与色彩 + +* 深蓝灰夜色; +* 小范围暖橙火光; +* 火光照亮泥地、靴子、背包边缘; +* 暗部保留; +* 远处不宜太亮。 + +注意: +火光不能把画面变成温馨露营片。 +它只能给人物一点点温度。 + +--- + +## 9. 与前后镜关系 + +### 前一镜 S-20 + +火近一点,照亮泥地和石头。 + +### 本镜 S-21 + +主角调整背包,借火光确认方向,继续上路。 + +### 后一镜 S-22 + +雨后微光、水洼反光,进入更开阔的继续感。 + +--- + +## 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* small real campfire; +* night roadside, muddy road; +* middle-aged traveler seen from side/back; +* adjusting old backpack; +* continuing forward, not camping; +* dark long coat, worn boots; +* restrained warm firelight; +* no group, no camping advertisement. + +这不是最终 Prompt,只是执行要点。 + +--- + +## 11. 验收关注点 + +审这一镜看 7 件事: + +1. 主角是不是和 S-05 的背影同一个人? +2. 背包、外套、靴子是否稳定? +3. 动作是否自然? +4. 火有没有太大、太温馨? +5. 有没有露营广告感? +6. 人物有没有停留成摆拍? +7. 这镜是否明确表达“火指引方向,而不是火成为终点”? + +--- + +# 镜头 S-25:我擦拭微尘 + +## 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 我擦拭微尘 + +时间位置约: + +> **1:59 - 3:24 主体副歌** + +这是副歌里一个身体动作镜头。 + +--- + +## 2. 本镜情绪功能 + +这一镜不是“擦脸耍帅”。 + +它表达的是: + +> 他不是干净地走过这条路。 +> 他身上有尘土,有疲惫,有走过泥泞后的痕迹。 + +这是一种克制的身体证明。 + +--- + +## 3. 核心画面 + +主角在风里短暂停下,用手擦过: + +* 脸侧; +* 额头; +* 外套肩部; +* 或衣领上的尘土。 + +画面可以只露半侧脸和手。 +手上有粗糙感,外套有尘土,风吹乱头发。 + +不要让动作变得戏剧化。 +它应该很短、很自然,像一个人走累后下意识做的动作。 + +--- + +## 4. 人物与镜头关系 + +### 人物可见度 + +* 局部; +* 半侧脸; +* 手部动作; +* 不完整正脸; +* 不直视镜头。 + +### 人物占比 + +建议: + +> **1/3 左右以内** + +这是人物局部动作镜头,可以稍近,但仍要克制。 + +--- + +## 5. 人物一致性锚点 + +这一镜验证的是近距离细节: + +1. **手部质感** + 有中年手的粗糙感,不精致,不干净得像广告手。 + +2. **脸侧轮廓** + 和 S-14、S-33 的脸部气质一致。 + +3. **胡茬** + 明显但自然,不时尚修饰。 + +4. **发型** + 深色、稍乱,被风吹过。 + +5. **外套材质** + 有尘土和使用痕迹,但不破烂。 + +--- + +## 6. 场景与视觉母题 + +调用: + +* 微尘; +* 手; +* 外套; +* 风; +* 中年行者; +* 身体疲惫。 + +这一镜是人物局部和环境痕迹的结合。 +它要让观众感到“这一路真的落在身体上”。 + +--- + +## 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 半侧构图; +* 手部动作在画面前景或中景; +* 脸不要完整正对; +* 背景虚化但能感到旷野或荒路; +* 不要干净影棚感。 + +画面应该像: + +> 镜头经过他身边时,捕捉到一个短暂的下意识动作。 + +--- + +## 8. 光线与色彩 + +* 风沙环境下的冷灰自然光; +* 或黄昏前的低饱和光; +* 皮肤不要过度磨皮; +* 尘土可见但不脏乱; +* 外套仍为深灰 / 暗卡其 / 煤黑。 + +不要: + +* 柔光人像; +* 皮肤精修; +* 明显摆拍; +* 过度煽情眼神。 + +--- + +## 9. 与前后镜关系 + +### 前一镜 S-24 + +人物在无名旷野里被世界吞没。 + +### 本镜 S-25 + +镜头拉近到身体痕迹,证明这条路真的磨过他。 + +### 后一镜 S-26 + +进入“仰望碎裂星海”的铺垫,空间再次打开。 + +这个节奏是: + +> 大远景压小人 → 局部看到尘土 → 再转向星海。 + +--- + +## 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* half-profile middle-aged traveler; +* hand wiping dust from face or coat; +* rough hand, stubble, messy dark hair; +* dark weathered long coat; +* wind and dust; +* natural candid action; +* not a portrait, not glamour. + +这不是最终 Prompt,只是执行要点。 + +--- + +## 11. 验收关注点 + +审这一镜看 7 件事: + +1. 动作是否自然? +2. 手是否有中年粗粝感? +3. 脸侧是否和 S-14 同一个人? +4. 胡茬和发型是否稳定? +5. 有没有擦脸写真感? +6. 尘土是否真实但不过脏? +7. 这个动作是否服务“走过风尘”,而不是服务人物好看? + +--- + +# 镜头 S-33:我咬紧牙关 + +## 1. 对应歌词 / 音乐段落 + +对应歌词: + +> 我咬紧牙关 + +时间位置约: + +> **1:59 - 3:24 主体副歌后段** + +这是全片少数可以接近人物情绪的镜头之一。 + +--- + +## 2. 本镜情绪功能 + +这一镜要表达的是内在用力。 + +不是怒吼。 +不是热血爆发。 +不是“我要逆袭”。 + +而是: + +> 他很累,但没有松。 +> 他咬住了自己,继续走。 + +这是一种中年人的硬撑,不是年轻人的燃爆。 + +--- + +## 3. 核心画面 + +半侧脸短近景。 + +画面中可以看到: + +* 被风吹乱的深色头发; +* 胡茬; +* 嘴角或下颌微微收紧; +* 眼神疲惫但有火; +* 深色外套领口; +* 背景是模糊的旷野、夜路或风沙。 + +注意: +“咬紧牙关”不一定要真的露牙。 +更建议用下颌、嘴角、眼神和呼吸状态表达。 + +--- + +## 4. 人物与镜头关系 + +### 人物可见度 + +* 半侧脸; +* 短近景; +* 不正面直视; +* 不做表演式表情。 + +### 人物占比 + +这一镜可以是人物占比最高的镜头之一: + +> **1/3 - 1/2** + +但必须短、克制,不能让它变成人物情绪写真。 + +--- + +## 5. 人物一致性锚点 + +这一镜是人物脸部一致性最重要的测试之一。 + +必须验证: + +1. **是否还是 S-14 的同一个人** +2. **年龄感是否稳定** +3. **胡茬是否一致** +4. **发型是否一致** +5. **眼神是否仍是“疲惫但有火”** +6. **有没有变成硬汉海报** +7. **有没有年轻化或帅化** + +这一镜最容易跑偏,因为它最接近人物脸。 + +--- + +## 6. 场景与视觉母题 + +调用: + +* 风沙; +* 中年行者; +* 胡茬; +* 咬紧牙关; +* 内在用力; +* 继续。 + +这不是风景镜头,但它必须仍属于风景叙事 MV。 +背景不能完全消失成影棚人像。 + +--- + +## 7. 构图方向 + +推荐: + +* 16:9 横屏; +* 半侧脸在画面一侧; +* 留出风、夜色或荒路背景; +* 头发和外套被风轻微影响; +* 不要正脸居中; +* 不要强眼神直视镜头。 + +镜头感觉应该是: + +> 人在路上经过一瞬间,镜头捕捉到他把自己撑住的表情。 + +--- + +## 8. 光线与色彩 + +可选两种光线: + +### 方案 A:冷风沙光 + +* 灰蓝自然光; +* 脸部暗部保留; +* 风尘感明显; +* 适合“咬紧牙关”的硬度。 + +### 方案 B:夜路微火边缘光 + +* 一点点暖火光扫过脸侧; +* 背景仍是深蓝黑; +* 情绪更温暖,但要避免煽情。 + +两种都可以,关键是不要: + +* 人像棚拍光; +* 英雄硬光; +* 精致电影男主光; +* 过度磨皮。 + +--- + +## 9. 与前后镜关系 + +### 前一镜 S-32 + +远处脚印或微弱火点暗示同行者。 + +### 本镜 S-33 + +镜头靠近主角,看到他内在用力的一瞬间。 + +### 后一镜 S-34 + +镜头落到靴跟和泥路,人物情绪立刻转回身体行动。 + +这个节奏很好: + +> 远方有人 → 他咬住自己 → 脚继续走。 + +--- + +## 10. 执行层要点 + +后续正式执行时,这一镜要强调: + +* half-profile close shot; +* middle-aged male traveler around 50; +* tired but determined; +* stubble, slightly messy dark hair; +* jaw tightened, restrained emotion; +* dark long coat collar visible; +* windy road background; +* no heroic poster pose; +* no glamour portrait. + +这不是最终 Prompt,只是执行要点。 + +--- + +## 11. 验收关注点 + +审这一镜看 8 件事: + +1. 是否和 S-14 是同一个人? +2. 是否保留 50 岁左右的中年感? +3. 胡茬、发型、外套是否稳定? +4. 有没有“疲惫但有火”? +5. 有没有表演过度? +6. 有没有英雄海报感? +7. 有没有精致男模化? +8. 看完这一镜,是否更相信他会继续走,而不是觉得他在摆拍? + +--- + +# 6. 这 5 镜的联动关系 + +这 5 个镜头组合起来,正好是一套人物一致性验证链: + +## S-05:先认背影 + +观众先认识他的背包、外套和肩背轮廓。 + +## S-21:再认动态 + +观众看到同一个人如何调整背包、继续走。 + +## S-14:确认脸部气质 + +观众第一次知道这个行者是怎样的中年人。 + +## S-25:确认身体痕迹 + +观众看到风尘、手、微尘,确认这条路落在身体上。 + +## S-33:确认精神状态 + +观众看到他最接近崩但没崩的一瞬间。 + +它们共同建立: + +> 他不是一个漂亮角色。 +> 他是一个被路磨过、但还在走的人。 + +--- + +# 7. 本轮测试的核心验收标准 + +这 5 镜测试完,不要只看单张美不美。 +重点看 5 个统一性问题: + +## 1. 背影和脸是不是同一个人? + +S-05 的背影,能不能自然连接到 S-14 和 S-33 的脸。 + +## 2. 服装有没有稳定? + +长外套、旧背包、靴子不能每镜都变。 + +## 3. 发型和胡茬有没有稳定? + +尤其 S-14、S-25、S-33 三个靠近脸的镜头。 + +## 4. 气质有没有稳定? + +始终是: + +> 沧桑摇滚,温暖疲惫,孤独但坚定。 + +不能一会儿流浪汉,一会儿男模,一会儿硬汉主角。 + +## 5. 人物有没有抢走 MV 的风景叙事基调? + +即便 S-33 是近一点的情绪镜头,也不能让全片变成人物写真。 + +--- + +# 8. 五镜最终判断标准 + +如果这 5 个镜头成立,人物系统基本就过了第一关。 + +因为它们覆盖了: + +* 背影; +* 侧背; +* 动作; +* 半侧脸; +* 手部; +* 情绪近点; +* 道具; +* 外套; +* 发型; +* 胡茬; +* 年龄; +* 气质。 + +也就是说,后面再做远景、城市、火光、水洼,就不容易“每张像不同人”。 + +本轮核心目标可以概括为一句: + +> **让观众不靠正脸,也能相信:这一路上走着的,始终是同一个人。**