mlsanityData to 7 Oct 2026

Rethinking matching-based few-shot action recognition

arXiv 2303.16084 ↗First posted 28 Mar 2023cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Few-shot action recognition, i.e. recognizing new action classes given only a few examples, benefits from incorporating temporal information. Prior work either encodes such information in the representation itself and learns classifiers at test time, or obtains frame-level features and performs pairwise temporal matching. We first evaluate a number of matching-based approaches using features from spatio-temporal backbones, a comparison missing from the literature, and show that the gap in performance between simple baselines and more complicated methods is significantly reduced. Inspired by this, we propose Chamfer++, a non-temporal matching function that achieves state-of-the-art results in few-shot action recognition. We show that, when starting from temporal features, our parameter-free and interpretable approach can outperform all other matching-based and classifier methods for one-shot action recognition on three common datasets without using temporal information in the matching stage. Project page: https://jbertrand89.github.io/matching-based-fsar

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 28 of its 38 references that are in the map. First posted on arXiv 28 Mar 2023.

Builds on

28 of 38 references.

References · 28

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  2. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  3. 2016Matching Networks for One Shot Learningcited by 2,246
  4. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  5. 2017A Closer Look at Spatiotemporal Convolutions for Action Recognitioncited by 1,139
  6. 2017The "something something" video database for learning and evaluating visual common sensecited by 1,092
  7. 2019A Closer Look at Few-shot Classificationcited by 679
  8. 2018Meta-Learning with Latent Embedding Optimizationcited by 476
  9. 2018Dynamic Few-Shot Visual Learning without Forgettingcited by 385
  10. 2018Probabilistic Model-Agnostic Meta-Learningcited by 254
  11. 2019SimpleShot: Revisiting Nearest-Neighbor Classification for Few-Shot Learningcited by 194
  12. 2018Recasting Gradient-Based Meta-Learning as Hierarchical Bayescited by 180
16 more
  1. 2020CrossTransformers: spatially-aware few-shot transfercited by 174
  2. 2019Few-Shot Video Classification via Temporal Alignmentcited by 124
  3. 2019D3TW: Discriminative Differentiable Dynamic Time Warping for Weakly Supervised Action Alignment and Segmentationcited by 83
  4. 2021Temporal-Relational CrossTransformers for Few-Shot Action Recognitioncited by 83
  5. 2019TARN: Temporal Attentive Relation Network for Few-Shot and Zero-Shot Action Recognitioncited by 71
  6. 2019Dense Classification and Implanting for Few-Shot Learningcited by 56
  7. 2021Spatio-temporal Relation Modeling for Few-shot Action Recognitioncited by 52
  8. 2022Hybrid Relation Guided Set Matching for Few-shot Action Recognitioncited by 49
  9. 2019ViSiL: Fine-grained Spatio-Temporal Video Similarity Learningcited by 41
  10. 2019ProtoGAN: Towards Few Shot Learning for Action Recognitioncited by 41
  11. 2021Drop-DTW: Aligning Common Signal Between Sequences While Dropping Outlierscited by 39
  12. 2021TA2N: Two-Stage Action Alignment Network for Few-shot Action Recognitioncited by 35
  13. 2022Compound Prototype Matching for Few-shot Action Recognitioncited by 35
  14. 2021Few-shot Action Recognition with Prototype-centered Attentive Learningcited by 28
  15. 2021A Closer Look at Few-Shot Video Classification: A New Baseline and Benchmarkcited by 16
  16. 2020Generalized Few-Shot Video Classification with Video Retrieval and Feature Generationcited by 6

Built on by

Not cited yet.