mlsanityData to 14 Oct 2026

Learning Spatio-Temporal Features with 3D Residual Networks for Action Recognition

arXiv 1708.07632 ↗First posted 25 Aug 2017cs.CVcited by 178

Not on the map. Why?

Abstract (arXiv)

Convolutional neural networks with spatio-temporal 3D kernels (3D CNNs) have an ability to directly extract spatio-temporal features from videos for action recognition. Although the 3D kernels tend to overfit because of a large number of their parameters, the 3D CNNs are greatly improved by using recent huge video databases. However, the architecture of 3D CNNs is relatively shallow against to the success of very deep neural networks in 2D-based CNNs, such as residual networks (ResNets). In this paper, we propose a 3D CNNs based on ResNets toward a better action representation. We describe the training procedure of our 3D ResNets in details. We experimentally evaluate the 3D ResNets on the ActivityNet and Kinetics datasets. The 3D ResNets trained on the Kinetics did not suffer from overfitting despite the large number of parameters of the model, and achieved better performance than relatively shallow networks, such as C3D. Our code and pretrained models (e.g. Kinetics and ActivityNet) are publicly available at https://github.com/kenshohara/3D-ResNets.

Key facts. Cited by 178 papers in the map (data through 14 Oct 2026). First posted on arXiv 25 Aug 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 37; 2022: 41; 2023: 36; 2024: 26; 2025: 26; 2026: 12.

178 papers.

Sort:

By area:

  1. 2023Video Understanding with Large Language Models: A Surveycited by 206
  2. 2022Frozen CLIP Models are Efficient Video Learnerscited by 182
  3. 2022FAST-VQA: Efficient End-to-end Video Quality Assessment with Fragment Samplingcited by 179
  4. 2021Exploring Temporal Coherence for More General Video Face Forgery Detectioncited by 159
  5. 2021TokenLearner: What Can 8 Learned Tokens Do for Images and Videos?cited by 127
  6. 2023Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlookcited by 110
  7. 2023TALL: Thumbnail Layout for Deepfake Video Detectioncited by 100
  8. 2021Blindly Assess Quality of In-the-Wild Videos via Quality-aware Pre-training and Motion Perceptioncited by 97
  9. 2022SVFormer: Semi-supervised Video Transformer for Action Recognitioncited by 68
  10. 2021FrameExit: Conditional Early Exiting for Efficient Video Recognitioncited by 59
  11. 2022Surgical-VQA: Visual Question Answering in Surgical Scenes using Transformercited by 59
  12. 2023CaptainCook4D: A Dataset for Understanding Errors in Procedural Activitiescited by 51
38 more
  1. 2021Semi-Supervised Action Recognition with Temporal Contrastive Learningcited by 50
  2. 2022Neighbourhood Representative Sampling for Efficient End-to-end Video Quality Assessmentcited by 50
  3. 2021Contrast and Mix: Temporal Contrastive Video Domain Adaptation with Background Mixingcited by 49
  4. 2023Exploring Deep Models for Practical Gait Recognitioncited by 46
  5. 2023Implicit Temporal Modeling with Learnable Alignment for Video Recognitioncited by 46
  6. 20214D-Net for Learned Multi-Modal Alignmentcited by 44
  7. 2022DisCoVQA: Temporal Distortion-Content Transformers for Video Quality Assessmentcited by 41
  8. 2022Delving into Sequential Patches for Deepfake Detectioncited by 41
  9. 2022S4ND: Modeling Images and Videos as Multidimensional Signals Using State Spacescited by 41
  10. 2021UNIK: A Unified Framework for Real-world Skeleton-based Action Recognitioncited by 35
  11. 2022Probabilistic Representations for Video Contrastive Learningcited by 34
  12. 2023Large Language Models are Clinical Reasoners: Reasoning-Aware Diagnosis Framework with Prompt-Generated Rationalescited by 33
  13. 2021Space-Time Crop & Attend: Improving Cross-modal Video Representation Learningcited by 31
  14. 2022Large-scale Robustness Analysis of Video Action Recognition Modelscited by 29
  15. 2021Coarse-Fine Networks for Temporal Activity Detection in Videoscited by 28
  16. 2022Source-free Video Domain Adaptation by Learning Temporal Consistency for Action Recognitioncited by 28
  17. 2021Enhancing Self-supervised Video Representation Learning via Multi-level Feature Optimizationcited by 27
  18. 2022A Deeper Dive Into What Deep Spatiotemporal Networks Encode: Quantifying Static vs. Dynamic Informationcited by 26
  19. 2024VideoMamba: Spatio-Temporal Selective State Space Modelcited by 25
  20. 2021Deepfake Detection Scheme Based on Vision Transformer and Distillationcited by 22
  21. 2022ViA: View-invariant Skeleton Action Representation Learning via Motion Retargetingcited by 22
  22. 2024PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour Recognitioncited by 22
  23. 2024Learning Spatiotemporal Inconsistency via Thumbnail Layout for Face Deepfake Detectioncited by 21
  24. 2021Multi-Source Video Domain Adaptation with Temporal Attentive Moment Alignmentcited by 20
  25. 2021Deep Depth from Focus with Differential Focus Volumecited by 20
  26. 2023Learning from Semantic Alignment between Unpaired Multiviews for Egocentric Video Recognitioncited by 20
  27. 2024SurGen: Text-Guided Diffusion Model for Surgical Video Generationcited by 20
  28. 2021Self-Conditioned Probabilistic Learning of Video Rescalingcited by 19
  29. 2023PECoP: Parameter Efficient Continual Pretraining for Action Quality Assessmentcited by 19
  30. 2023AutoLabel: CLIP-based framework for Open-set Video Domain Adaptationcited by 18
  31. 2024Image anomaly detection and prediction scheme based on SSA optimized ResNet50-BiGRU modelcited by 18
  32. 2023Vision-Language Models can Identify Distracted Driver Behavior from Naturalistic Videoscited by 17
  33. 2021EAN: Event Adaptive Network for Enhanced Action Recognitioncited by 16
  34. 2023Learning Situation Hyper-Graphs for Video Question Answeringcited by 16
  35. 2023Temporal Coherent Test-Time Optimization for Robust Video Classificationcited by 15
  36. 2023LAC: Latent Action Composition for Skeleton-based Action Segmentationcited by 15
  37. 2024OpenGait: A Comprehensive Benchmark Study for Gait Recognition towards Better Practicalitycited by 15
  38. 2021Metadata Normalizationcited by 14

50 most cited of 178 All 178 (JSON)