Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions

1 code implementation • 29 Mar 2024 • Runhao Zeng, Xiaoyong Chen, Jiaming Liang, Huisi Wu, Guangzhong Cao, Yong Guo

In this paper, we extensively analyze the robustness of seven leading TAD methods and obtain some interesting findings: 1) Existing methods are particularly vulnerable to temporal corruptions, and end-to-end methods are often more susceptible than those with a pre-trained feature extractor; 2) Vulnerability mainly comes from localization error rather than classification error; 3) When corruptions occur in the middle of an action instance, TAD models tend to yield the largest performance drop.

Action Detection Benchmarking