TPT-for-VideoQA icon indicating copy to clipboard operation
TPT-for-VideoQA copied to clipboard

Published 20 hours ago • verified publisher icon Trunpm

Video feature extraction model

Open WeiYangBin opened this issue 3 years ago • 1 comments

Hello, what model is used in the second step to extract features， can you provide some information

WeiYangBin avatar Nov 15 '21 08:11 WeiYangBin

A pre-trained ResNet152 model is used to extract the feature. ref. https://drive.google.com/open?id=1xbYbZ7rpyjftI_KCk6YuL-XrfQDz7Yd4 @WeiYangBin

Trunpm avatar Nov 17 '21 10:11 Trunpm