Metadata-Version: 2.4
Name: reshot
Version: 0.4.0
Summary: ReShot — copy the shot, not the actors. Video → depth map for Seedance 2.0/2.5 reference video, MiniMax H3 Fun ControlNet and Wan VACE. Open-sourced by Maosika 猫斯卡 (www.maosika.com), the AI short-drama production system.
Author: Maosika 猫斯卡
License: Apache-2.0
Project-URL: Homepage, https://github.com/maosika-ai/reshot
Project-URL: Documentation, https://github.com/maosika-ai/reshot/blob/main/docs/USAGE.md
Project-URL: ComfyUI nodes, https://github.com/maosika-ai/ComfyUI-ReShot
Project-URL: Changelog, https://github.com/maosika-ai/reshot/blob/main/CHANGELOG.md
Project-URL: Maosika, https://www.maosika.com
Keywords: video,depth,controlnet,video-generation,depth-map,depth-anything,seedance,minimax-h3,ai-short-drama,maosika
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Multimedia :: Video
Classifier: Topic :: Multimedia :: Graphics
Classifier: Intended Audience :: End Users/Desktop
Classifier: Operating System :: OS Independent
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
License-File: NOTICE
Requires-Dist: torch>=2.1
Requires-Dist: torchvision>=0.16
Requires-Dist: numpy>=1.24
Requires-Dist: opencv-python-headless>=4.8
Requires-Dist: einops>=0.4
Requires-Dist: easydict
Requires-Dist: tqdm
Requires-Dist: huggingface_hub>=0.20
Provides-Extra: ffmpeg
Requires-Dist: imageio-ffmpeg>=0.4.7; extra == "ffmpeg"
Provides-Extra: dev
Requires-Dist: pytest>=8; extra == "dev"
Requires-Dist: ruff; extra == "dev"
Dynamic: license-file

# ReShot

**Copy the shot, not the actors.**

ReShot turns a reference video into a depth-map video (near = white, far = black), so Seedance, MiniMax H3 or Wan can repeat its choreography and camera moves with your own characters.

```bash
pip install reshot
reshot reference.mp4 -o depth.mp4 --target seedance
```

- Presets for **Seedance 2.0 / 2.5** reference video (`--target seedance`), **MiniMax H3** (`h3`), **Wan 2.1 VACE** (`wan`).
- `--quality fast` (default; the model sees 644×364 for 16:9, ~3 GB VRAM) or `full` (924×518, ~11 GB).
- Depth normalised once over the whole clip — no flicker; frames picked by timestamp; sizes cropped to the model's grid, never padded.
- Batch a folder in one model load: `reshot clips/*.mp4 -o depth/`.
- Python API: `from reshot import RunConfig, run`.
- Apache-2.0, including the default model (Video Depth Anything Small).
- ComfyUI nodes: https://github.com/maosika-ai/ComfyUI-ReShot

Full guide, demo, measured performance and the prompts that made the demo takes: **https://github.com/maosika-ai/reshot**

Open-sourced by [Maosika 猫斯卡](https://www.maosika.com), the AI short-drama production system — from a one-line idea to episodic script, character sheets, scene images and every shot rendered with Seedance and MiniMax H3. https://www.maosika.com
