How can I run inference with my own video?

e-apostolidis / PGL-SUM

A PyTorch Implementation of PGL-SUM from "Combining Global and Local Attention with Positional Encoding for Video Summarization" (IEEE ISM 2021)

Other

81 stars 32 forks source link

How can I run inference with my own video? #3

Closed jet981217 closed 2 years ago

jet981217 commented 2 years ago

How can I extract the features for my own video so that I can put it in inference.py?

mpalaourg commented 2 years ago

Hello @jet981217, this repo doesn't support inferencing for your own video. It is possible if you get the frame_features via GoogleNet and by running -for a specific model- the inference.py script.

However, we have a web service here, which lets you submit videos in various formats and generate summaries for use in various social media channels. I hope this helps!

jet981217 commented 2 years ago

@mpalaourg Thank you so much for the swift answer! This will help a lot for me!