Commit Graph

7 Commits

Author SHA1 Message Date
69a4aebfe7 Add auto_pyr_level toggle to select pyramid level by resolution
When enabled (default), automatically picks the optimal pyr_level
based on input height: <540p=3, 540p=5, 1080p=6, 4K=7.
When disabled, uses the manual pyr_level value.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:51:29 +01:00
4e6f9eb896 Respect user's pyr_level setting at all resolutions
Previously the user's pyr_level was overridden for >=540p content.
Now the setting is always used, with the tooltip recommending values
per resolution instead.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:50:29 +01:00
ffde07a89a Add tooltips to all node inputs
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:49:27 +01:00
2c84b47590 Auto-install cupy and gdown on first node load
Detects missing deps when ComfyUI imports the node and installs
them automatically via pip. No manual steps needed in Docker.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:43:29 +01:00
d44885252f Write detected cupy variant into requirements.txt at install time
install.py now updates requirements.txt with the correct cupy-cuda
package matching PyTorch's CUDA version before running pip install.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:42:23 +01:00
471a299027 Fix: lazy import cupy to allow node loading without cupy installed
cupy was imported at module level, causing ComfyUI to fail loading
the node entirely if cupy wasn't installed yet. Now deferred to
first actual CUDA kernel call, with a clear error message.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:35:11 +01:00
db64fc195a Initial commit: ComfyUI BIM-VFI node for video frame interpolation
Wraps BiM-VFI (CVPR 2025) as a ComfyUI custom node for long video
frame interpolation with memory-safe sequential processing.

- LoadBIMVFIModel: checkpoint loader with auto-download from Google Drive
- BIMVFIInterpolate: 2x/4x/8x recursive interpolation with per-pair
  GPU processing, configurable VRAM management (all_on_gpu for high-VRAM
  setups), progress bar, and backwarp cache clearing
- Vendored inference-only architecture from KAIST-VICLab/BiM-VFI
- Auto-detection of CUDA version for cupy installation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:26:49 +01:00