mlsanityData to 14 Oct 2026

Otter: Mitigating Background Distractions of Wide-Angle Few-Shot Action Recognition with Enhanced RWKV

arXiv 2511.06741 ↗First posted 10 Nov 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Wide-angle videos in few-shot action recognition (FSAR) effectively express actions within specific scenarios. However, without a global understanding of both subjects and background, recognizing actions in such samples remains challenging because of the background distractions. Receptance Weighted Key Value (RWKV), which learns interaction between various dimensions, shows promise for global modeling. While directly applying RWKV to wide-angle FSAR may fail to highlight subjects due to excessive background information. Additionally, temporal relation degraded by frames with similar backgrounds is difficult to reconstruct, further impacting performance. Therefore, we design the CompOund SegmenTation and Temporal REconstructing RWKV (Otter). Specifically, the Compound Segmentation Module~(CSM) is devised to segment and emphasize key patches in each frame, effectively highlighting subjects against background information. The Temporal Reconstruction Module (TRM) is incorporated into the temporal-enhanced prototype construction to enable bidirectional scanning, allowing better reconstruct temporal relation. Furthermore, a regular prototype is combined with the temporal-enhanced prototype to simultaneously enhance subject emphasis and temporal modeling, improving wide-angle FSAR performance. Extensive experiments on benchmarks such as SSv2, Kinetics, UCF101, and HMDB51 demonstrate that Otter achieves state-of-the-art performance. Extra evaluation on the VideoBadminton dataset further validates the superiority of Otter in wide-angle FSAR.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 34 of its 48 references that are in the map. First posted on arXiv 10 Nov 2025.

Builds on

34 of 48 references.

References · 34

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  2. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  3. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  4. 2017The Kinetics Human Action Video Datasetcited by 1,909
  5. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,143
  6. 2017The "something something" video database for learning and evaluating visual common sensecited by 1,103
  7. 2016Temporal Segment Networks: Towards Good Practices for Deep Action Recognitioncited by 902
  8. 2023RWKV: Reinventing RNNs for the Transformer Eracited by 827
  9. 2019Meta-Learning with Implicit Gradientscited by 525
  10. 2018Meta-Learning with Latent Embedding Optimizationcited by 476
  11. 2018TADAM: Task dependent adaptive metric for improved few-shot learningcited by 408
  12. 2023OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perceptioncited by 221
22 more
  1. 2018Low-Shot Learning from Imaginary Datacited by 181
  2. 2024Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrencecited by 159
  3. 2018Task-Agnostic Meta-Learning for Few-shot Learningcited by 151
  4. 2020Adversarial Feature Hallucination Networks for Few-Shot Learningcited by 110
  5. 2024Vision-RWKV: Efficient and Scalable Visual Perception with RWKV-Like Architecturescited by 98
  6. 2021Temporal-Relational CrossTransformers for Few-Shot Action Recognitioncited by 84
  7. 2019Image Deformation Meta-Networks for One-Shot Learningcited by 56
  8. 2020Depth Guided Adaptive Meta-Fusion Network for Few-shot Video Recognitioncited by 50
  9. 2024Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Modelscited by 41
  10. 2021CTRL-C: Camera calibration TRansformer with Line-Classificationcited by 33
  11. 2021Learning Implicit Temporal Alignment for Few-shot Video Classificationcited by 27
  12. 2022C3-STISR: Scene Text Image Super-resolution with Triple Cluescited by 26
  13. 2023Boosting Few-shot Action Recognition with Graph-guided Hybrid Matchingcited by 24
  14. 2024RWKV-CLIP: A Robust Vision-Language Representation Learnercited by 23
  15. 2024PointRWKV: Efficient RWKV-Like Model for Hierarchical Point Cloud Learningcited by 21
  16. 2023Deep Learning for Camera Calibration and Beyond: A Surveycited by 20
  17. 2024Continuous-Multiple Image Outpainting in One-Step via Positional Query and A Diffusion-based Approachcited by 18
  18. 2020Cooperative Bi-path Metric for Few-shot Learningcited by 15
  19. 2022A Deep Perceptual Measure for Lens and Camera Calibrationcited by 7
  20. 2021Correcting Face Distortion in Wide-Angle Videoscited by 4
  21. 2024Benchmarking Badminton Action Recognition with a New Fine-Grained Datasetcited by 2
  22. 2024Manta: Enhancing Mamba for Few-Shot Action Recognition of Long Sub-Sequencecited by 2

Built on by

Not cited yet.