# AI use cases for Comfy UI H3 Face Refine (Carasibana/ComfyUI-H3-FaceRefine)

Repo: [Carasibana/ComfyUI-H3-FaceRefine](https://github.com/Carasibana/ComfyUI-H3-FaceRefine) | Use cases: 3 | Page: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/

A ComfyUI custom node set that tracks small faces in MiniMax H3 video, crops to each face, refines it with H3, and composites it back into the clip.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Refine small faces in a MiniMax H3 video: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/refine-small-faces-in-h3-video/
- Pick the right face in a multi-person video: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/pick-the-right-face-in-multi-person-clip/
- Check the face crop before a full H3 render: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/check-face-crop-before-full-render/

## How to refine small faces in a MiniMax H3 video with Comfy UI H3 Face Refine

Repo: [Carasibana/ComfyUI-H3-FaceRefine](https://github.com/Carasibana/ComfyUI-H3-FaceRefine) | Difficulty: Intermediate | Time: 3 hours | Basis: From the README

Track each face per frame, crop to it, re-generate it with H3, and stitch it back into the clip.

**Who it is for:** A video creator whose MiniMax H3 clip has smeared faces because the head fills only a small part of the frame.

**The problem:** H3 renders faces poorly when the head is a small fraction of the frame, and that stays true at 720p and above.

**The outcome:** A composited video where the small face is re-generated from a normalized crop and the rest of the frame keeps its original pixels.

### Steps

1. Run `git clone https://github.com/Carasibana/ComfyUI-H3-FaceRefine.git` inside ComfyUI/custom_nodes/.
2. Restart ComfyUI and confirm the nodes appear under MiniMax H3/Face Refine.
3. Place face_yolov8m.pt in models/ultralytics/bbox/, and the H3 diffusion model, Qwen3-VL text encoder, and VAEs in their folders.
4. Load your clip, choose the subject with the select setting, and connect it to H3 Face Track + Crop.
5. Send the crop through the H3 node, composite the result back, and save the video with a node such as VHS_VideoCombine.

### Prompt to paste into your agent

```text
Using the repo Carasibana/ComfyUI-H3-FaceRefine, build a ComfyUI graph that refines the small face of <subject description> in <video path> and writes a composited video to <output path>. Verify that the tracker report shows the subject found on most frames and lists any lost frame ranges.
```

### Caveat

The H3 models and the face detector are not bundled with the pack, so downloading and placing them is the slowest part of setup.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/refine-small-faces-in-h3-video/

## How to pick the right face in a multi-person video with Comfy UI H3 Face Refine

Repo: [Carasibana/ComfyUI-H3-FaceRefine](https://github.com/Carasibana/ComfyUI-H3-FaceRefine) | Difficulty: Intermediate | Time: 2 hours | Basis: From the README

Choose the subject per shot by eye so the tracker stays on the right person across hard cuts.

**Who it is for:** An editor with a clip showing several people, where the tracker might follow someone other than the intended subject.

**The problem:** A hard cut renumbers every face, and with several people in shot the tracker can settle on the wrong face.

**The outcome:** A subject chosen once per shot before the graph runs, so the crop follows the intended person through the whole clip.

### Steps

1. Run `git clone https://github.com/Carasibana/ComfyUI-H3-FaceRefine.git` inside ComfyUI/custom_nodes/.
2. Restart ComfyUI and add the H3 Load Video + Face Select node to your graph.
3. Open Pick faces and choose the subject for each shot.
4. Connect the face_pick output to H3 Face Track + Crop and keep cut_detection set to auto (pyscenedetect).
5. Run the graph and read the report to confirm each shot used the face you picked.

### Prompt to paste into your agent

```text
Using the repo Carasibana/ComfyUI-H3-FaceRefine, set up the H3 Load Video + Face Select node for <video path> so the subject is chosen per shot. Return the list of shots and the chosen face for each, and verify that every shot has a subject assigned.
```

### Caveat

Cut detection depends on PySceneDetect, and if it fails to import the video is treated as a single shot.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/pick-the-right-face-in-multi-person-clip/

## How to check the face crop before a full H3 render with Comfy UI H3 Face Refine

Repo: [Carasibana/ComfyUI-H3-FaceRefine](https://github.com/Carasibana/ComfyUI-H3-FaceRefine) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README

Use preview output to confirm the crop sits still on the subject before spending a full H3 render.

**Who it is for:** A VFX artist or creator who wants to avoid wasting a long H3 render on a crop that wobbles or jumps to another person.

**The problem:** A crop that moves or boils on a moving subject ruins the refined face, and a full render is expensive to repeat.

**The outcome:** A short check that shows whether the box holds still on the subject and which frames the subject was lost, before the full run.

### Steps

1. Run `git clone https://github.com/Carasibana/ComfyUI-H3-FaceRefine.git` inside ComfyUI/custom_nodes/.
2. Restart ComfyUI and open H3 Face Track + Crop with the preview output enabled.
3. Check the preview for the crop staying steady on the subject, and note any frames drawn red.
4. Read the lost line and the dropout runs in the report to see which frame ranges lost the subject.
5. Use Preview coordinates to confirm where an X and Y value lands, then start the full render.

### Prompt to paste into your agent

```text
Using the repo Carasibana/ComfyUI-H3-FaceRefine, run the H3 Face Track + Crop preview on <video path> without starting a full H3 render. Report the face= value, the lost ranges, and any frames drawn red, and verify that the subject is found on the frames you expect.
```

### Caveat

Frames where the subject is lost are left unrefined and keep their original pixels, so a clean preview does not mean every frame was refined.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/carasibana-comfyui-h3-facerefine/check-face-crop-before-full-render/