Repository navigation
Releases
v0.1.1 — Clearer settings and control-map comparison
Advanced settings are grouped into Video, Pose, Depth and Canny. Only selected extractors show their options, and hidden settings keep their values. Model choices sit alongside their related controls.
- Refreshed English and Korean node screenshots.
- Added a synchronized three-second Pose / Depth / Canny preview and 24 FPS MP4.
- Added prominent MiniMax H3 Fun ControlNet connection instructions.
- Verified group visibility, retained settings, minimum node size, both languages and playback in the browser; six settings tests pass.
Choose either the EN or KO archive. They differ only in the default interface language. Replace the existing node files and refresh the browser with Ctrl+F5. Do not install both copies.
Video Control Studio v0.1.0
Video Control Studio extracts Pose, Depth and Canny control sequences from video inside ComfyUI.
- Synchronized A/B playback, draggable comparison divider and frame stepping.
- Segment selection, FPS conversion, resize and workflow presets.
- One-frame tests and segment extraction run only the node and its inputs, even in a connected generation graph.
- Clean IMAGE and silent VIDEO outputs.
- Controls fit the minimum node size without internal scrolling.
- English and Korean UI, illustrated guides and four canvas workflows: extraction/export and H3 Pose ControlNet generation in both languages.
Install one ZIP under ComfyUI/custom_nodes, then restart ComfyUI and refresh the browser. EN and KO archives differ only in their default UI language. Control output and its choices remain English in both versions.
Canny needs no model. Pose requires SDPose Wholebody; Depth requires Depth Anything 3. Model weights are not bundled. See the README for required ComfyUI features and known limitations.