# Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions

arXiv: https://arxiv.org/abs/2403.20254 · HTML: https://mlsanity.com/paper/2403.20254/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 71 of its 89 references that are in the map.
- First posted on arXiv 29 Mar 2024.

## Builds on (most cited first, 30 of 71)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization](https://mlsanity.com/paper/2006.16241/) (2020)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty](https://mlsanity.com/paper/1912.02781/) (2019)
- [AutoAugment: Learning Augmentation Policies from Data](https://mlsanity.com/paper/1805.09501/) (2018)
- [VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking](https://mlsanity.com/paper/2303.16727/) (2023)
- [Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey](https://mlsanity.com/paper/1801.00553/) (2018)
- [Improving robustness against common corruptions by covariate shift adaptation](https://mlsanity.com/paper/2006.16971/) (2020)
- [Wild Patterns: Ten Years After the Rise of Adversarial Machine Learning](https://mlsanity.com/paper/1712.03141/) (2017)
- [Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations](https://mlsanity.com/paper/2202.07800/) (2022)
- [Rethinking Spatial Dimensions of Vision Transformers](https://mlsanity.com/paper/2103.16302/) (2021)
- [Adversarial Examples: Attacks and Defenses for Deep Learning](https://mlsanity.com/paper/1712.07107/) (2017)
- [BMN: Boundary-Matching Network for Temporal Action Proposal Generation](https://mlsanity.com/paper/1907.09702/) (2019)
- [ActionFormer: Localizing Moments of Actions with Transformers](https://mlsanity.com/paper/2202.07925/) (2022)
- [A Fourier Perspective on Model Robustness in Computer Vision](https://mlsanity.com/paper/1906.08988/) (2019)
- [Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples](https://mlsanity.com/paper/2010.03593/) (2020)
- [Temporal Action Detection with Structured Segment Networks](https://mlsanity.com/paper/1704.06228/) (2017)
- [G-TAD: Sub-Graph Localization for Temporal Action Detection](https://mlsanity.com/paper/1911.11462/) (2019)
- [Understanding Robustness of Transformers for Image Classification](https://mlsanity.com/paper/2103.14586/) (2021)
- [Adversarial Attacks and Defenses in Images, Graphs and Text: A Review](https://mlsanity.com/paper/1909.08072/) (2019)
- [Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shift](https://mlsanity.com/paper/2006.10963/) (2020)
- [Graph Convolutional Networks for Temporal Action Localization](https://mlsanity.com/paper/1909.03252/) (2019)
- [Vision Transformers are Robust Learners](https://mlsanity.com/paper/2105.07581/) (2021)
- [Bag of Tricks for Adversarial Training](https://mlsanity.com/paper/2010.00467/) (2020)

## Built on by (most cited first, 12 of 12)

- [Training-free Detection of Generated Videos via Spatial-Temporal Likelihoods](https://mlsanity.com/paper/2603.15026/) (2026)
- [PoseBench: Benchmarking the Robustness of Pose Estimation Models under Corruptions](https://mlsanity.com/paper/2406.14367/) (2024)
- [Towards Dependable Optical Flow: Benchmarks for Real-World Spatiotemporal Corruptions](https://mlsanity.com/paper/2411.14865/) (2024)
- [RoHOI: Robustness Benchmark for Human-Object Interaction Detection](https://mlsanity.com/paper/2507.09111/) (2025)
- [Human-Centric Open-Future Task Discovery: Formulation, Benchmark, and Scalable Tree-Based Search](https://mlsanity.com/paper/2511.18929/) (2025)
- [Investigating and Enhancing the Robustness of Large Multimodal Models Against Temporal Inconsistency](https://mlsanity.com/paper/2505.14405/) (2025)
- [Are Video Reasoning Models Ready to Go Outside?](https://mlsanity.com/paper/2603.10652/) (2026)
- [RobustGait: Robustness Analysis for Appearance Based Gait Recognition](https://mlsanity.com/paper/2511.13065/) (2025)
- [OZ-TAL: Online Zero-Shot Temporal Action Localization](https://mlsanity.com/paper/2605.09976/) (2026)
- [Suppress and Diversify: Refining Robust Pathways for Corruption Robustness](https://mlsanity.com/paper/2608.06712/) (2026)
- [Reliable Egocentric Action Anticipation via Temporal Reliability Suppression and Compositional Graph Decoding](https://mlsanity.com/paper/2609.13293/) (2026)
- [Are Vision-Language-Action Models Robust to One-Step Observation Perturbations?](https://mlsanity.com/paper/2609.32550/) (2026)

## Abstract (arXiv)

Temporal action detection (TAD) aims to locate action positions and recognize action categories in long-term untrimmed videos. Although many methods have achieved promising results, their robustness has not been thoroughly studied. In practice, we observe that temporal information in videos can be occasionally corrupted, such as missing or blurred frames. Interestingly, existing methods often incur a significant performance drop even if only one frame is affected. To formally evaluate the robustness, we establish two temporal corruption robustness benchmarks, namely THUMOS14-C and ActivityNet-v1.3-C. 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. Besides building a benchmark, we further develop a simple but effective robust training method to defend against temporal corruptions, through the FrameDrop augmentation and Temporal-Robust Consistency loss. Remarkably, our approach not only improves robustness but also yields promising improvements on clean data. We believe that this study will serve as a benchmark for future research in robust video analysis. Source code and models are available at https://github.com/Alvin-Zeng/temporal-robustness-benchmark.