Giters

V-Sense / ACTION-Net

Official PyTorch implementation of ACTION-Net: Multipath Excitation for Action Recognition (CVPR'21)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Question about TSM result

gd2016229035 opened this issue · comments

commented

In Jester dataset, TSM result in Table 1 of your paper is 94.4 (ResNet50 8frame 3crop 10clip),
But TSM result in origin paper(Table 1) is 97.0 (ResNet50 8frame full-resolution 2clips ).
I test the TSM by myself from the TSM official public code and it can also reach 97.0.
How cause this Gap between 94.4 and 97.0?

commented

In Jester dataset, TSM result in Table 1 of your paper is 94.4 (ResNet50 8frame 3crop 10clip),
But TSM result in origin paper(Table 1) is 97.0 (ResNet50 8frame full-resolution 2clips ).
I test the TSM by myself from the TSM official public code and it can also reach 97.0.
How cause this Gap between 94.4 and 97.0?

Hi there, thanks for your question. We just directly use the result in STM paper in Table 2. I guess it might be caused by using different training augmentation strategies.