指南首页 / 案例拆解 / 渡口 / 视频与表演
示范案例 · 06

《渡口》视频与表演

对照实际输入、结构化输出和质量检查,理解这一阶段怎样连接上下游。

10H3 生成任务
2已完成视频段
29s已完成素材时长
8已完成关键帧

每个生成段都有对齐指令、逐镜动作、原文对白、整体声景和非剧情音乐字段。

REELBENCH 已完成视频

可直接播放的生成结果

E01-01 · 雾中奔向渡口 · 15 秒
E01-02 · 登船与戒备 · 14 秒
E01-01 与 E01-02 已生成并可在页面播放;其余 8 个任务尚未生成。
SOURCE DATA

H3 投产清单

当前页面可直接查看全部 10 个生成任务的输入、输出与资产状态。

展开原始数据
[
  {
    "segment": "E01-01",
    "seconds": 15,
    "cuts": 4,
    "cutStarts": [
      0,
      3,
      6,
      10
    ],
    "prompt": "h3-package/E01-01/prompt.md",
    "pictures": [
      "h3-package/E01-01/f1.png",
      "h3-package/E01-01/f2.png",
      "h3-package/E01-01/f3.png",
      "h3-package/E01-01/f4.png"
    ],
    "missing": [
      "h3-package/E01-01/f1.png",
      "h3-package/E01-01/f2.png",
      "h3-package/E01-01/f3.png",
      "h3-package/E01-01/f4.png"
    ]
  },
  {
    "segment": "E01-02",
    "seconds": 14,
    "cuts": 4,
    "cutStarts": [
      0,
      3,
      6,
      10
    ],
    "prompt": "h3-package/E01-02/prompt.md",
    "pictures": [
      "h3-package/E01-02/f1.png",
      "h3-package/E01-02/f2.png",
      "h3-package/E01-02/f3.png",
      "h3-package/E01-02/f4.png"
    ],
    "missing": [
      "h3-package/E01-02/f1.png",
      "h3-package/E01-02/f2.png",
      "h3-package/E01-02/f3.png",
      "h3-package/E01-02/f4.png"
    ]
  },
  {
    "segment": "E01-03",
    "seconds": 12,
    "cuts": 4,
    "cutStarts": [
      0,
      4,
      7,
      10
    ],
    "prompt": "h3-package/E01-03/prompt.md",
    "pictures": [
      "h3-package/E01-03/f1.png",
      "h3-package/E01-03/f2.png",
      "h3-package/E01-03/f3.png",
      "h3-package/E01-03/f4.png"
    ],
    "missing": [
      "h3-package/E01-03/f1.png",
      "h3-package/E01-03/f2.png",
      "h3-package/E01-03/f3.png",
      "h3-package/E01-03/f4.png"
    ]
  },
  {
    "segment": "E01-04",
    "seconds": 11,
    "cuts": 3,
    "cutStarts": [
      0,
      3,
      8
    ],
    "prompt": "h3-package/E01-04/prompt.md",
    "pictures": [
      "h3-package/E01-04/f1.png",
      "h3-package/E01-04/f2.png",
      "h3-package/E01-04/f3.png"
    ],
    "missing": [
      "h3-package/E01-04/f1.png",
      "h3-package/E01-04/f2.png",
      "h3-package/E01-04/f3.png"
    ]
  },
  {
    "segment": "E01-05",
    "seconds": 13,
    "cuts": 3,
    "cutStarts": [
      0,
      5,
      8
    ],
    "prompt": "h3-package/E01-05/prompt.md",
    "pictures": [
      "h3-package/E01-05/f1.png",
      "h3-package/E01-05/f2.png",
      "h3-package/E01-05/f3.png"
    ],
    "missing": [
      "h3-package/E01-05/f1.png",
      "h3-package/E01-05/f2.png",
      "h3-package/E01-05/f3.png"
    ]
  },
  {
    "segment": "E01-06",
    "seconds": 12,
    "cuts": 4,
    "cutStarts": [
      0,
      2,
      5,
      9
    ],
    "prompt": "h3-package/E01-06/prompt.md",
    "pictures": [
      "h3-package/E01-06/f1.png",
      "h3-package/E01-06/f2.png",
      "h3-package/E01-06/f3.png",
      "h3-package/E01-06/f4.png"
    ],
    "missing": [
      "h3-package/E01-06/f1.png",
      "h3-package/E01-06/f2.png",
      "h3-package/E01-06/f3.png",
      "h3-package/E01-06/f4.png"
    ]
  },
  {
    "segment": "E01-07",
    "seconds": 11,
    "cuts": 3,
    "cutStarts": [
      0,
      3,
      7
    ],
    "prompt": "h3-package/E01-07/prompt.md",
    "pictures": [
      "h3-package/E01-07/f1.png",
      "h3-package/E01-07/f2.png",
      "h3-package/E01-07/f3.png"
    ],
    "missing": [
      "h3-package/E01-07/f1.png",
      "h3-package/E01-07/f2.png",
      "h3-package/E01-07/f3.png"
    ]
  },
  {
    "segment": "E01-08",
    "seconds": 14,
    "cuts": 4,
    "cutStarts": [
      0,
      4,
      7,
      11
    ],
    "prompt": "h3-package/E01-08/prompt.md",
    "pictures": [
      "h3-package/E01-08/f1.png",
      "h3-package/E01-08/f2.png",
      "h3-package/E01-08/f3.png",
      "h3-package/E01-08/f4.png"
    ],
    "missing": [
      "h3-package/E01-08/f1.png",
      "h3-package/E01-08/f2.png",
      "h3-package/E01-08/f3.png",
      "h3-package/E01-08/f4.png"
    ]
  },
  {
    "segment": "E01-09",
    "seconds": 9,
    "cuts": 3,
    "cutStarts": [
      0,
      3,
      6
    ],
    "prompt": "h3-package/E01-09/prompt.md",
    "pictures": [
      "h3-package/E01-09/f1.png",
      "h3-package/E01-09/f2.png",
      "h3-package/E01-09/f3.png"
    ],
    "missing": [
      "h3-package/E01-09/f1.png",
      "h3-package/E01-09/f2.png",
      "h3-package/E01-09/f3.png"
    ]
  },
  {
    "segment": "E01-10",
    "seconds": 8,
    "cuts": 2,
    "cutStarts": [
      0,
      5
    ],
    "prompt": "h3-package/E01-10/prompt.md",
    "pictures": [
      "h3-package/E01-10/f1.png",
      "h3-package/E01-10/f2.png"
    ],
    "missing": [
      "h3-package/E01-10/f1.png",
      "h3-package/E01-10/f2.png"
    ]
  }
]
ALL PRODUCTION PROMPTS

10 段完整 H3 投产提示词

E01-01 · 15 秒 · 4 个切点
# E01-01 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 3.00 秒)
- Picture 3 = f3.png(钉 6.00 秒)
- Picture 4 = f4.png(钉 10.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 3.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 6.00-second mark of the target video; Picture 4 (from Shot 4) aligns with the 10.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The cold open follows <Picture 1>: a young woman in a plain qipao runs through thick fog along a muddy riverside path, reeds and puddles at her feet, an old leather suitcase hugged tight against her chest; a tracking shot follows her from behind, the pier not yet in sight as the fog swallows her silhouette.
[Shot 2] At 00:03.000, the camera cuts to the extreme wide view of <Picture 2>: the run ends and the wooden pier finally emerges from the fog, the old ferryman squatting at the bow of the ferry as a faint silhouette with his pipe ember glowing through the mist; the camera holds a static shot while distant watchman clappers knock, muffled as if wrapped in cotton.
[Shot 3] At 00:06.000, the shot cuts to the framing of <Picture 3>: the ferryman stands at the bow with one hand cupped to his mouth, the camera holds a static shot, and the weathered old boatman with a loud, coarse, unhurried voice (S1) calls toward the shore: <d>[Chinese] 上船喽——过河的抓紧,雾要变天。</d>
[Shot 4] At 00:10.000, the shot cuts to <Picture 4> and a tracking shot follows the middle-aged peddler with a bright, breathless voice (S2) hurrying along the gangplank under a loaded shoulder pole, shouting: <d>[Chinese] 等等二爷我!这雾天赶船,命都得跑掉半条。</d>

overall_soundscape: Rapid running footsteps splash along a wet riverside path through fog-damped silence, water laps softly against the pilings, wooden clappers knock twice in the far distance, and rattling baskets close in later along the gangplank.

non_diegetic_music: Sustained low strings at a very slow tempo, barely above silence.
E01-02 · 14 秒 · 4 个切点
# E01-02 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 3.00 秒)
- Picture 3 = f3.png(钉 6.00 秒)
- Picture 4 = f4.png(钉 10.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 3.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 6.00-second mark of the target video; Picture 4 (from Shot 4) aligns with the 10.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The gunwale view of <Picture 1> holds as the middle-aged peddler steps aboard with his loaded shoulder pole and the old ferryman lends a hand, the two of them setting the load down into the cabin together; the camera holds a static shot as the baskets settle.
[Shot 2] At 00:03.000, the camera cuts to <Picture 2>: on board, the ferryman stands by the cabin entrance facing into the cabin, pipe still in his teeth; the camera holds a static shot as the weathered old boatman with a coarse, dry-humored voice (S1) says toward the inside: <d>[Chinese] 胡二爷,你这担子比人金贵。</d>
[Shot 3] At 00:06.000, the shot cuts to <Picture 3>: seated in the cabin beside his baskets, the grinning peddler pats them, and the middle-aged peddler with a bright, boastful voice (S2) laughs back over a static shot: <d>[Chinese] 那可不——吃饭的家伙,比命金贵。</d>
[Shot 4] At 00:10.000, the shot cuts to <Picture 4>: a young woman in a plain qipao steps onto the wet gangplank with an old leather suitcase hugged to her chest, and a slow push in with small amplitude closes on her whitened knuckles around the case edge.

overall_soundscape: Baskets creak softly as the load settles onto the cabin floor with a light chime of small brass bells, wood knocks gently, and slow careful footsteps creak along the wet gangplank.

non_diegetic_music: N/A
E01-03 · 12 秒 · 4 个切点
# E01-03 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 4.00 秒)
- Picture 3 = f3.png(钉 7.00 秒)
- Picture 4 = f4.png(钉 10.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 4.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 7.00-second mark of the target video; Picture 4 (from Shot 4) aligns with the 10.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. Holding the two-figure composition of <Picture 1> near the bow, the old ferryman keeps coiling rope without lifting his eyes; the camera holds a static shot as the weathered old boatman with a flat, indifferent voice (S1) mutters: <d>[Chinese] 箱子沉就搁船头,掉河里我可不捞。</d>
[Shot 2] At 00:04.000, the camera cuts to the close framing of <Picture 2>: the young woman tightens both arms around the suitcase, a gentle push in with small amplitude, and the quiet young woman with a soft but unyielding voice (S2) answers: <d>[Chinese] 不劳烦。它跟我。</d>
[Shot 3] At 00:07.000, the shot cuts to <Picture 3>: the old ferryman gives her one brief glance without a word, then pulls the gangplank aboard; the camera holds a static shot.
[Shot 4] At 00:10.000, the shot cuts to <Picture 4> and a slow pull out opens across the dark water as (S1) plants the pole and calls: <d>[Chinese] 坐稳——开船。</d> — the ferry slides forward and the pier lantern dissolves into fog.

overall_soundscape: Rope fibers rasp over the gunwale, the bamboo pole knocks once and dips with a heavy push, and the gangplank planks are dragged aboard.

non_diegetic_music: Sparse low strings at a slow tempo fade in for a few seconds and out again.
E01-04 · 11 秒 · 3 个切点
# E01-04 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 3.00 秒)
- Picture 3 = f3.png(钉 8.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 3.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 8.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The dim cabin of <Picture 1> holds: goods baskets hog half a bench while the man in the long dark coat leans against the wall, right hand buried in his coat pocket, completely still; the camera holds a static shot as the cabin rocks gently.
[Shot 2] At 00:03.000, the camera cuts to the grinning peddler of <Picture 2> leaning toward the aisle, static shot, and the middle-aged peddler with a bright, over-familiar voice (S1) asks: <d>[Chinese] 姑娘头回走这条水路吧?我一看一个准。</d>
[Shot 3] At 00:08.000, the shot cuts to <Picture 3> and a tracking shot follows the young woman crossing to the farthest corner seat without answering, settling the old leather suitcase on her knees and folding both arms around it.

overall_soundscape: The hull creaks in a slow rhythm over water sliding beneath the boards, basket straps rustle, and footsteps knock softly across the deck before a bench creaks.

non_diegetic_music: N/A
E01-05 · 13 秒 · 3 个切点
# E01-05 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 5.00 秒)
- Picture 3 = f3.png(钉 8.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 5.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 8.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The close framing of <Picture 1> holds on the peddler half risen from his bench; the camera holds a static shot as the peddler with a warm, pushy voice (S1) presses: <d>[Chinese] 城里亲戚接站,还是投店?二爷我熟门熟路。</d>
[Shot 2] At 00:05.000, the camera cuts to the reverse close-up of <Picture 2>: the young woman keeps her eyes lowered over a static shot, and the quiet young woman with a flat, guarded voice (S2) replies: <d>[Chinese] 借过就好,不敢劳动。</d>
[Shot 3] At 00:08.000, the shot cuts to <Picture 3>: his hand starts toward the suitcase on her knees while a slow push in with small amplitude tightens the frame, and (S1) grins wider: <d>[Chinese] 嗨,出门在外别客气——箱子给我,搁上头稳当。</d>

overall_soundscape: A bench scrapes as he rises, clothes rustle in the narrow cabin, and the hull keeps its slow creak over passing water.

non_diegetic_music: N/A
E01-06 · 12 秒 · 4 个切点
# E01-06 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 2.00 秒)
- Picture 3 = f3.png(钉 5.00 秒)
- Picture 4 = f4.png(钉 9.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 2.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 5.00-second mark of the target video; Picture 4 (from Shot 4) aligns with the 9.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The insert of <Picture 1> lands hard: the young woman's hand slams down on the suitcase lid, pinning the brass clasp under her fingers — fast and hard; the camera holds a static shot on the impact.
[Shot 2] At 00:02.000, the camera cuts to <Picture 2>: the cabin chatter dies, the peddler's reaching hand hangs frozen mid-air and his grin stiffens; static shot.
[Shot 3] At 00:05.000, the shot cuts to <Picture 3>, through the opening toward the bow, static shot, and the weathered old boatman with a slow, even voice (S1) says without stopping his pole work: <d>[Chinese] 船上有个规矩——不问来路,不碰行李。</d>
[Shot 4] At 00:09.000, the shot cuts to <Picture 4>: the peddler pulls his hand back and bows his head twice over a static shot, and the peddler with a deflated, placating voice (S2) mutters: <d>[Chinese] 是是,二爷我多嘴,多嘴。</d>

overall_soundscape: A flat sharp slap of palm on leather cuts through the cabin followed by a sudden hush, then the bamboo pole dips and drips in a steady cycle at the bow.

non_diegetic_music: N/A
E01-07 · 11 秒 · 3 个切点
# E01-07 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 3.00 秒)
- Picture 3 = f3.png(钉 7.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 3.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 7.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The mid-river view of <Picture 1> holds: fog hugging the water into one gray sheet, the ferry a dark sliver, the ferryman a calm vertical figure at the bow; static shot while the peddler's restless silhouette stirs in the cabin.
[Shot 2] At 00:03.000, the camera cuts to <Picture 2>, static shot, and the peddler with a fidgety, talk-filling voice (S1) calls toward the bow: <d>[Chinese] 老周,今儿这雾邪性,走得完吗?</d>
[Shot 3] At 00:07.000, the shot cuts to the bow of <Picture 3>: without turning, over a static shot with the pole rising and falling, the weathered old boatman with an unhurried, granite-steady voice (S2) answers: <d>[Chinese] 走了四十年,没有走不完的河。</d>

overall_soundscape: The pole enters and leaves the water in slow even strokes, fog-muffled ripples spread along the hull, and a basket strap creaks as the peddler shifts his weight.

non_diegetic_music: N/A
E01-08 · 14 秒 · 4 个切点
# E01-08 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 4.00 秒)
- Picture 3 = f3.png(钉 7.00 秒)
- Picture 4 = f4.png(钉 11.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 4.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 7.00-second mark of the target video; Picture 4 (from Shot 4) aligns with the 11.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. Keeping the cabin composition of <Picture 1>, the peddler turns to the man in the long dark coat; static shot, and the peddler with a probing, cheerful voice (S1) tries: <d>[Chinese] 这位先生面生得很,也是去县城?</d>
[Shot 2] At 00:04.000, the camera cuts to the close framing of <Picture 2>: the man in the dark coat gives a single faint nod, eyes never leaving the fog; static shot.
[Shot 3] At 00:07.000, the shot cuts to <Picture 3>: the peddler shrugs theatrically to the whole cabin, static shot, and the same cheerful voice (S1) sighs: <d>[Chinese] 得,一船的闷葫芦,就我一个人长嘴。</d>
[Shot 4] At 00:11.000, the shot cuts to <Picture 4>: from inside the cabin the old ferryman's back shakes once with a short laugh he does not turn around for; static shot while water ticks under the hull like something counting down.

overall_soundscape: The hull groans softly on a slow swell, a short huff of breath accompanies the shrug, then one dry laugh from the bow over water knocking in a steady patient rhythm.

non_diegetic_music: N/A
E01-09 · 9 秒 · 3 个切点
# E01-09 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 3.00 秒)
- Picture 3 = f3.png(钉 6.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 3.00-second mark of the target video; Picture 3 (from Shot 3) aligns with the 6.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. Holding the framing of <Picture 1>, the fabric over the dark coat pocket tightens as the hidden right hand clenches once; a slow push in with small amplitude from chest height toward the pocket.
[Shot 2] At 00:03.000, the camera cuts to <Picture 2>: his face half in shadow, lips barely moving, and the man in the dark coat with a low, restrained voice (S1) murmurs almost to himself: <d>[Chinese] 这条河,今晚怕是不太平。</d> static shot.
[Shot 3] At 00:06.000, the shot cuts to the hushed cabin of <Picture 3>: a muffled steam whistle sounds once deep in the fog off screen and no one answers it; the camera holds a static shot on the stillness.

overall_soundscape: Coat fabric creaks faintly under the clenching hand, water passes below in a low wash, and a distant muffled steam whistle rolls once through the fog.

non_diegetic_music: A single sustained low cello tone at a very slow tempo, fading before the end.
E01-10 · 8 秒 · 2 个切点
# E01-10 · H3 提示词

首帧 = **f1.png**。图片按 Picture 序号挂载:

- Picture 1 = f1.png(**首帧**,钉 0.00 秒)
- Picture 2 = f2.png(钉 5.00 秒)

---

How the reference pictures align with the target video — Picture 1 (from Shot 1) aligns with the 0.00-second mark of the target video; Picture 2 (from Shot 2) aligns with the 5.00-second mark of the target video.

integrated_multimodal_description:
[Shot 1] Cinematic, live-action, cold gray-green palette. The extreme close framing of <Picture 1> holds on the green-tarnished brass clasp, the young woman's arms wrapped around the case edge; a slow push in with small amplitude as fog creeps across the frame, and the young woman with a cold, steady voice (S1) says in an off-screen voiceover: <d>[Chinese] 谁都别碰它。到了县城,它就是我全部的话。</d> while her lips remain completely closed.
[Shot 2] At 00:05.000, the camera cuts to <Picture 2>: the clasp glints once, and in the blurred distance the pier lantern shrinks to a bean of yellow light and dies; the camera holds a static shot, leaving the dull glint as the last bright point in the frame.

overall_soundscape: Water slides under the hull beneath the fading whistle's echo, leather creaks under tightening arms, and the cabin holds its hush as the distant light goes out.

non_diegetic_music: Sparse low strings and a single soft piano note at a slow tempo, thinning to silence as the light dies.
这页有用吗?
哪里不好?

收到,谢谢。我们会根据反馈改这一页。