目录 / shipinduanjushengcheng
shipinduanjushengcheng
Create short videos by transferring a reference video's motion/actions onto the person(s) in a photo — "让照片里的人做出视频里的动作". Takes a photo (the subject) plus a reference video (the motion source), analyzes the video's actions, then generates a new video where the photo's subject performs those actions. Supports full-body dance/sports and multi-person choreography. Use when the user provides (or wants to provide) a photo + a video and asks to: make the person in the photo do the actions/dance/movements from the video; transfer motion from a video to a photo; 让照片里的人做视频里的动作; 照片+视频生成视频; 动作迁移/动作模仿; 换人跳舞; 模仿视频动作; motion transfer; video reenactment; photo-to-video with reference motion. Also triggers on "用这张照片模仿这段视频的动作" or similar.
这是模型对公开材料的总结,不是官方声明,请以原链内容为准。
存档时间线
| 版本 | 存档时间 | 内容哈希 | 内容 |
|---|---|---|---|
| v1 | 2026-11-12 20:32 | 355f1e86 | 可取 |
版本索引永久保留;内容副本只保留最新 1 份,更早版本仅留索引与哈希(存档时间线的证据链不会因此断裂)。
提交举报 / 纠错
侵权举报经核验成立后,我们会即时下线该条目并删除已存的内容副本。