Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Proposal-Free Temporal Action Detection via Global Segmentation Mask Learning

About

Existing temporal action detection (TAD) methods rely on generating an overwhelmingly large number of proposals per video. This leads to complex model designs due to proposal generation and/or per-proposal action instance evaluation and the resultant high computational cost. In this work, for the first time, we propose a proposal-free Temporal Action detection model with Global Segmentation mask (TAGS). Our core idea is to learn a global segmentation mask of each action instance jointly at the full video length. The TAGS model differs significantly from the conventional proposal-based methods by focusing on global temporal representation learning to directly detect local start and end points of action instances without proposals. Further, by modeling TAD holistically rather than locally at the individual proposal level, TAGS needs a much simpler model architecture with lower computational cost. Extensive experiments show that despite its simpler design, TAGS outperforms existing TAD methods, achieving new state-of-the-art performance on two benchmarks. Importantly, it is ~ 20x faster to train and ~1.6x more efficient for inference. Our PyTorch implementation of TAGS is available at https://github.com/sauradip/TAGS .

Sauradip Nag, Xiatian Zhu, Yi-Zhe Song, Tao Xiang• 2022

Related benchmarks

TaskDatasetResultRank
Temporal Action DetectionTHUMOS-14 (test)
mAP@tIoU=0.557
330
Temporal Action DetectionActivityNet v1.3 (val)
mAP@0.556.3
185
Temporal Action DetectionActivityNet 1.3
mAP@0.556.3
93
Temporal Action LocalizationTHUMOS 2014
mAP@0.3068.6
93
Temporal Action DetectionActivityNet 1.3 (test)
Average mAP36.5
80
Temporal Action DetectionTHUMOS 14
mAP@0.368.6
71
Showing 6 of 6 rows

Other info

Code

Follow for update