Using pre-trained model for video feature extraction

facebookresearch / SlowFast

PySlowFast: video understanding codebase from FAIR for reproducing state-of-the-art video models.

Apache License 2.0

6.52k stars 1.2k forks source link

Using pre-trained model for video feature extraction #394

Open arpitkumar1412 opened 3 years ago

arpitkumar1412 commented 3 years ago

Hi! I wanted to use a pre-trained-X3D model for extracting features from a video by removing the final classification layer from the model. I have already extracted frames from the video but could not find support for running pre-trained models on custom set of frames. TIA

wanghao14 commented 3 years ago

@arpitkumar1412 Hi, I encounter the same question as you. Had you solved it?