Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up

vapoursynth version for DRBA

License

NotificationsYou must be signed in to change notification settings

routineLife1/VS-DRBA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Distance Ratio Based Adjuster for Animeinterp, based onhttps://github.com/routineLife1/DRBA andhttps://github.com/HolyWu/vs-rife.

This project is modified fromHolyWu/vs-rife and achieves nearly the same interpolation quality as the originalDRBA project.

With TensorRT integration, it achieves a 400% acceleration, enabling real-time playback on high-performance NVIDIA GPUs.

Dependencies

trt requires additional packages:

To install the latest stable version of PyTorch, Torch-TensorRT and cupy, run:

pip install -U packaging setuptools wheelpip install -U torch torchvision torch_tensorrt --index-url https://download.pytorch.org/whl/cu126 --extra-index-url https://pypi.nvidia.compip install -U cupy-cuda12x

Installation

pip install -U vsdrba==1.0.1

If you want to download all models at once, runpython -m vsdrba. If you prefer to only download the model youspecified at first run, setauto_download=True indrba_rife().

Usage

fromvsdrbaimportdrba_riferet=drba_rife(clip)

See__init__.py for the description of the parameters.

Benchmarks

modelscaleoshardwarearchfps 720fps 1080vram 720vram 1080backendverified outputbatchlevelstreamsthreadsonnxonnxslim / onnxsimonnx shapetrtexec shapeprecisionusage
rife 4.26 heavy2xLinux3070laptop / 12400rife (4.26)119531.6gb3.4gbtrt 10.8, torch 20241231+cu126, torch_trt 20250102+cu126 (holywu vsrife)yes, works15-8---staticRGBHrife(clip, trt=True, trt_static_shape=True, model="4.26.heavy", trt_optimization_level=5, sc=False)
drba_rife 4.26 heavy2xLinux3070laptop / 12400drba_rife (4.26)158701.7gb3.7gbtrt 10.8, torch 20241231+cu126, torch_trt 20250102+cu126 (routineLife1 vsdrba)yes, works15-8---staticRGBHrife(clip, trt=True, trt_static_shape=True, model="4.26.heavy", trt_optimization_level=5, sc=False)

[8]ページ先頭

©2009-2025 Movatter.jp