https://openreview.net/forum?id=1jrYSOG7DR
Revealing the unseen: Benchmarking video action recognition under occlusion | OpenReview
In this work, we study the effect of occlusion on video action recognition. To facilitate this study, we propose three benchmark datasets and experiment with...
video action recognitionthe unseenrevealingbenchmarkingocclusion
https://openreview.net/forum?id=CIoSZ_HKHS7
AIM: Adapting Image Models for Efficient Video Action Recognition | OpenReview
We propose a new method to adapt frozen image pre-trained model for efficient video action recognition
video action recognitionimage modelsaimadaptingefficient
https://deepai.org/publication/no-frame-left-behind-full-video-action-recognition
No frame left behind: Full Video Action Recognition | DeepAI
Mar 29, 2021 - 03/29/21 - Not all video frames are equally informative for recognizing an action. It is computationally infeasible to train deep networks on...
video action recognitionno frameleft behindfulldeepai
https://openreview.net/forum?id=8ApJf7S-I5&referrer=%5Bthe%20profile%20of%20Feixiang%20He%5D(%2Fprofile%3Fid%3D~Feixiang_He1)
Local fusion networks with chained residual pooling for video action recognition | OpenReview
video action recognitionfusion networks
https://openreview.net/forum?id=SXt3pGxOpB&referrer=%5Bthe%20profile%20of%20Deepti%20Ghadiyaram%5D(%2Fprofile%3Fid%3D~Deepti_Ghadiyaram2)
Large-Scale Weakly-Supervised Pre-Training for Video Action Recognition. | OpenReview
Current fully-supervised video datasets consist of only a few hundred thousand videos and fewer than a thousand domain-specific labels. This hinders the...
video action recognitionlarge scalepre trainingsupervised