mlsanityData to 14 Oct 2026

A Large-Scale Study on Video Action Dataset Condensation

arXiv 2412.21197 ↗First posted 30 Dec 2024cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Recently, dataset condensation has made significant progress in the image domain. Unlike images, videos possess an additional temporal dimension, which harbors considerable redundant information, making condensation even more crucial. However, video dataset condensation still remains an underexplored area. We aim to bridge this gap by providing a large-scale study with systematic design and fair comparison. Specifically, our work delves into three key aspects to provide valuable empirical insights: (1) temporal processing of video data, (2) the evaluation protocol for video dataset condensation, and (3) adaptation of condensation algorithms to the space-time domain. From this study, we derive several intriguing observations: (i) labeling methods greatly influence condensation performance, (ii) simple sliding-window sampling is effective for temporal processing, and (iii) dataset distillation methods perform better in challenging scenarios, while sample selection methods excel in easier ones. Furthermore, we propose a unified evaluation protocol for the fair comparison of different condensation algorithms and achieve state-of-the-art results on four widely-used action recognition datasets: HMDB51, UCF101, SSv2 and K400. Our code is available at https://github.com/MCG-NJU/Video-DC.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 43 of its 53 references that are in the map. First posted on arXiv 30 Dec 2024.

Builds on

43 of 53 references.

References · 43

  1. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  2. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  3. 2018SlowFast Networks for Video Recognitioncited by 1,867
  4. 2021Is Space-Time Attention All You Need for Video Understanding?cited by 1,620
  5. 2022VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Trainingcited by 1,322
  6. 2017A Closer Look at Spatiotemporal Convolutions for Action Recognitioncited by 1,143
  7. 2018An Empirical Study of Example Forgetting during Deep Neural Network Learningcited by 678
  8. 2021Deep Learning on a Data Diet: Finding Important Examples Early in Trainingcited by 533
  9. 2020Dataset Condensation with Gradient Matchingcited by 516
  10. 2022Dataset Distillation by Matching Training Trajectoriescited by 442
  11. 2021Dataset Condensation with Distribution Matchingcited by 365
  12. 2021Dataset Condensation with Differentiable Siamese Augmentationcited by 291
31 more
  1. 2021GRAD-MATCH: Gradient Matching based Data Subset Selection for Efficient Deep Model Trainingcited by 274
  2. 2020Dataset Meta-Learning from Kernel Ridge-Regressioncited by 230
  3. 2020GLISTER: Generalization based Data Subset Selection for Efficient and Robust Learningcited by 226
  4. 2021Dataset Distillation with Infinitely Wide Convolutional Networkscited by 218
  5. 2022Dataset Condensation via Efficient Synthetic-Data Parameterizationcited by 188
  6. 2022Scaling Up Dataset Distillation to ImageNet-1K with Constant Memorycited by 181
  7. 2022Dataset Distillation using Neural Feature Regressioncited by 177
  8. 2021Graph Condensation for Graph Neural Networkscited by 150
  9. 2022DeepCore: A Comprehensive Library for Coreset Selection in Deep Learningcited by 147
  10. 2023Squeeze, Recover and Relabel: Dataset Condensation at ImageNet Scale From A New Perspectivecited by 138
  11. 2023Generalizing Dataset Distillation via Deep Generative Priorcited by 126
  12. 2023Towards Lossless Dataset Distillation via Difficulty-Aligned Trajectory Matchingcited by 118
  13. 2022Dataset Distillation via Factorizationcited by 115
  14. 2022Efficient Dataset Distillation Using Random Feature Approximationcited by 111
  15. 2023On the Diversity and Realism of Distilled Dataset: An Efficient Dataset Distillation Paradigmcited by 109
  16. 2019Generative Teaching Networks: Accelerating Neural Architecture Search by Learning to Generate Synthetic Training Datacited by 104
  17. 2022Remember the Past: Distilling Datasets into Addressable Memories for Neural Networkscited by 104
  18. 2020Characterizing Structural Regularities of Labeled Data in Overparameterized Modelscited by 93
  19. 2023Generalized Large-Scale Data Condensation via Various Backbone and Statistical Matchingcited by 69
  20. 2022DC-BENCH: Dataset Condensation Benchmarkcited by 67
  21. 2023Does Graph Distillation See Like Vision Dataset Counterpart?cited by 45
  22. 2024Elucidating the Design Space of Dataset Condensationcited by 40
  23. 2024A Label is Worth a Thousand Images in Dataset Distillationcited by 31
  24. 2024SelMatch: Effectively Scaling Up Dataset Distillation via Selection-Based Initialization and Partial Updates by Trajectory Matchingcited by 22
  25. 2023Dancing with Still Images: Video Distillation via Static-Dynamic Disentanglementcited by 21
  26. 2024Towards Stable and Storage-efficient Dataset Distillation: Matching Convexified Trajectorycited by 21
  27. 2023Vision-Language Dataset Distillationcited by 19
  28. 2024Emphasizing Discriminative Features for Dataset Distillation in Complex Scenarioscited by 18
  29. 2024Dataset Condensation for Time Series Classification via Dual Domain Matchingcited by 9
  30. 2024Two Trades is not Baffled: Condensing Graph via Crafting Rational Gradient Matchingcited by 8
  31. 2024CondTSF: One-line Plugin of Dataset Condensation for Time Series Forecastingcited by 6

Built on by

Cited by year

Citing papers per year: 2025: 6.

6 papers.

  1. 2025Retrieval Augmented Generation and Understanding in Vision: A Survey and New Outlookcited by 32
  2. 2025VideoITG: Multimodal Video Understanding with Instructed Temporal Groundingcited by 29
  3. 2025The Evolution of Dataset Distillation: Toward Scalable and Generalizable Solutionscited by 26
  4. 2025CovMatch: Cross-Covariance Guided Multimodal Dataset Distillation with Trainable Text Encodercited by 3
  5. 2025VideoCompressa: Data-Efficient Video Understanding via Joint Temporal Compression and Spatial Reconstructioncited by 3
  6. 2025Temporal Saliency-Guided Distillation: A Scalable Framework for Distilling Video Datasetscited by 1