mlsanityData to 14 Oct 2026

MoExDA: Domain Adaptation for Edge-based Action Recognition

arXiv 2508.02981 ↗First posted 5 Aug 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Modern action recognition models suffer from static bias, leading to reduced generalization performance. In this paper, we propose MoExDA, a lightweight domain adaptation between RGB and edge information using edge frames in addition to RGB frames to counter the static bias issue. Experiments demonstrate that the proposed method effectively suppresses static bias with a lower computational cost, allowing for more robust action recognition than previous approaches.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 45 references that are in the map. First posted on arXiv 5 Aug 2025.

Builds on

33 of 45 references.

References · 33

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  4. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  5. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  6. 2017The Kinetics Human Action Video Datasetcited by 1,909
  7. 2018SlowFast Networks for Video Recognitioncited by 1,867
  8. 2021Is Space-Time Attention All You Need for Video Understanding?cited by 1,620
  9. 2021ViViT: A Video Vision Transformercited by 1,620
  10. 2018Group Normalizationcited by 1,524
  11. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,198
  12. 2021Video Swin Transformercited by 1,091
21 more
  1. 2020X3D: Expanding Architectures for Efficient Video Recognitioncited by 637
  2. 2021VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understandingcited by 602
  3. 2017Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?cited by 551
  4. 2021ActionCLIP: A New Paradigm for Video Action Recognitioncited by 320
  5. 2018Human Action Recognition and Prediction: A Surveycited by 227
  6. 2022Fine-tuned CLIP Models are Efficient Video Learnerscited by 179
  7. 2019Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognitioncited by 117
  8. 2020On Feature Normalization and Data Augmentationcited by 71
  9. 2020Removing the Background by Adding the Background: Towards Background Robust Self-supervised Video Representation Learningcited by 66
  10. 2022Video Transformers: A Surveycited by 63
  11. 2020VideoMix: Rethinking Data Augmentation for Video Classificationcited by 55
  12. 2019Mimetics: Towards Understanding Human Actions Out of Contextcited by 32
  13. 2022Mitigating and Evaluating Static Bias of Action Representations in the Background and the Foregroundcited by 27
  14. 2016Human Action Recognition without Humancited by 17
  15. 2022Is Appearance Free Action Recognition Possible?cited by 16
  16. 2020Video Action Understandingcited by 13
  17. 2022ObjectMix: Data Augmentation by Copy-Pasting Objects in Videos for Action Recognitioncited by 7
  18. 2022Mitigating Representation Bias in Action Recognition: Algorithms and Benchmarkscited by 7
  19. 2023DEVIAS: Learning Disentangled Video Representations of Action and Scenecited by 7
  20. 2025Can masking background and object reduce static bias for zero-shot action recognition?cited by 6
  21. 2023S3Aug: Segmentation, Sampling, and Shift for Action Recognitioncited by 3

Built on by

Not cited yet.