mlsanityData to 7 Oct 2026

Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation

arXiv 2103.14304 ↗First posted 26 Mar 2021cs.CVcited by 94

Not on the map. Why?

Abstract (arXiv)

Despite the great progress in 3D human pose estimation from videos, it is still an open problem to take full advantage of a redundant 2D pose sequence to learn representative representations for generating one 3D pose. To this end, we propose an improved Transformer-based architecture, called Strided Transformer, which simply and effectively lifts a long sequence of 2D joint locations to a single 3D pose. Specifically, a Vanilla Transformer Encoder (VTE) is adopted to model long-range dependencies of 2D pose sequences. To reduce the redundancy of the sequence, fully-connected layers in the feed-forward network of VTE are replaced with strided convolutions to progressively shrink the sequence length and aggregate information from local contexts. The modified VTE is termed as Strided Transformer Encoder (STE), which is built upon the outputs of VTE. STE not only effectively aggregates long-range information to a single-vector representation in a hierarchical global and local fashion, but also significantly reduces the computation cost. Furthermore, a full-to-single supervision scheme is designed at both full sequence and single target frame scales applied to the outputs of VTE and STE, respectively. This scheme imposes extra temporal smoothness constraints in conjunction with the single target frame supervision and hence helps produce smoother and more accurate 3D poses. The proposed Strided Transformer is evaluated on two challenging benchmark datasets, Human3.6M and HumanEva-I, and achieves state-of-the-art results with fewer parameters. Code and models are available at urlhttps://github.com/Vegetebird/StridedTransformer-Pose3D.

Key facts. Cited by 94 papers in the map (data through 7 Oct 2026). Builds on 36 of its 66 references that are in the map. First posted on arXiv 26 Mar 2021.

Builds on

36 of 66 references.

References · 36

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2020End-to-End Object Detection with Transformerscited by 7,384
  4. 2017Mask R-CNNcited by 7,259
  5. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,410
  6. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,307
  7. 2021Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNetcited by 1,096
  8. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  9. 2020Efficient Transformers: A Surveycited by 784
  10. 2021TransReID: Transformer-based Object Re-Identificationcited by 432
  11. 2020End-to-End Human Pose and Mesh Reconstruction with Transformerscited by 417
  12. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
24 more
  1. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 401
  2. 2017Cascaded Pyramid Network for Multi-Person Pose Estimationcited by 341
  3. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 271
  4. 2017Integral Human Pose Regressioncited by 255
  5. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 220
  6. 2019Semantic Graph Convolutional Networks for 3D Human Pose Regressioncited by 204
  7. 2020Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processingcited by 161
  8. 2020Lite Transformer with Long-Short Range Attentioncited by 127
  9. 2018Ordinal Depth Supervision for 3D Human Pose Estimationcited by 107
  10. 2019Self-Supervised Learning of 3D Human Pose using Multi-view Geometrycited by 98
  11. 2020SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approachcited by 94
  12. 2017Exploiting temporal information for 3D pose estimationcited by 89
  13. 2020ConvBERT: Improving BERT with Span-based Dynamic Convolutioncited by 88
  14. 2020Motion Guided 3D Pose Estimation from Videoscited by 85
  15. 2021Trear: Transformer-based RGB-D Egocentric Action Recognitioncited by 39
  16. 2018Memory Attention Networks for Skeleton-based Action Recognitioncited by 36
  17. 2019Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimationcited by 30
  18. 2018A generalizable approach for multi-view 3D human pose regressioncited by 26
  19. 2019Feature Boosting Network For 3D Pose Estimationcited by 26
  20. 2019Weakly-Supervised Discovery of Geometry-Aware Representation for 3D Human Pose Estimationcited by 23
  21. 2018Depth Pooling Based Large-scale 3D Action Recognition with Convolutional Neural Networkscited by 19
  22. 2018Rethinking Pose in 3D: Multi-stage Refinement and Recovery for Markerless Motion Capturecited by 17
  23. 2020Transformer Guided Geometry Model for Flow-Based Unsupervised Visual Odometrycited by 5
  24. 2017Robust 3D Action Recognition through Sampling Local Appearances and Global Distributionscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 18; 2023: 29; 2024: 17; 2025: 20; 2026: 5.

94 papers.

Sort:

By area:

  1. 2022MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Videocited by 158
  2. 2021MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimationcited by 151
  3. 2023PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimationcited by 91
  4. 2023Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregationcited by 81
  5. 2022P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimationcited by 80
  6. 2023MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Networkcited by 62
  7. 2023GLA-GCN: Global-local Adaptive Graph Convolutional Network for 3D Human Pose Estimation from Monocular Videocited by 46
  8. 2022ViTCoD: Vision Transformer Acceleration via Dedicated Algorithm and Accelerator Co-Designcited by 41
  9. 2022Tracking Objects as Pixel-wise Distributionscited by 36
  10. 2022Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformerscited by 33
  11. 2024KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimationcited by 33
  12. 2022DiffuPose: Monocular 3D Human Pose Estimation via Denoising Diffusion Probabilistic Modelcited by 29
38 more
  1. 2023Co-Evolution of Pose and Mesh for 3D Human Body Estimation from Videocited by 25
  2. 2023HDFormer: High-order Directed Transformer for 3D Human Pose Estimationcited by 23
  3. 20233D Human Pose Perception from Egocentric Stereo Videoscited by 23
  4. 2022GraphMLP: A Graph MLP-Like Architecture for 3D Human Pose Estimationcited by 21
  5. 20223D Vision with Transformers: A Surveycited by 21
  6. 2021Adaptive Multi-view and Temporal Fusing Transformer for 3D Human Pose Estimationcited by 19
  7. 2022Contrastive Learning from Spatio-Temporal Mixed Skeleton Sequences for Self-Supervised Skeleton-Based Action Recognitioncited by 17
  8. 2022Occluded Human Body Capture with Self-Supervised Spatial-Temporal Motion Priorcited by 15
  9. 2023Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimationcited by 15
  10. 2025AthletePose3D: A Benchmark Dataset for 3D Human Pose Estimation and Kinematic Validation in Athletic Movementscited by 14
  11. 2021Weakly-supervised 3D Human Pose Estimation with Cross-view U-shaped Graph Convolutional Networkcited by 13
  12. 2023Monocular 3D Human Pose Estimation for Sports Broadcasts using Partial Sports Field Registrationcited by 13
  13. 2021FLEX: Extrinsic Parameters-free Multi-view 3D Human Motion Reconstructioncited by 11
  14. 2022CrossFormer: Cross Spatio-Temporal Transformer for 3D Human Pose Estimationcited by 11
  15. 2023Regular Splitting Graph Network for 3D Human Pose Estimationcited by 11
  16. 2023A Single 2D Pose with Context is Worth Hundreds for 3D Human Pose Estimationcited by 11
  17. 2022FeatER: An Efficient Network for Human Reconstruction via Feature Map-Based TransformERcited by 9
  18. 2023Weakly Supervised Multi-Modal 3D Human Body Pose Estimation for Autonomous Drivingcited by 8
  19. 2022A Dual-Masked Auto-Encoder for Robust Motion Capture with Spatial-Temporal Skeletal Token Completioncited by 7
  20. 2023OAMatcher: An Overlapping Areas-based Network for Accurate Local Feature Matchingcited by 7
  21. 2025Bring Your Rear Cameras for Egocentric 3D Human Pose Estimationcited by 7
  22. 2022Pyramid Fusion Transformer for Semantic Segmentationcited by 6
  23. 2023Human Pose-based Estimation, Tracking and Action Recognition with Deep Learning: A Surveycited by 6
  24. 2025Towards Robust and Realistic Human Pose Estimation via WiFi Signalscited by 5
  25. 2025H2OT: Hierarchical Hourglass Tokenizer for Efficient Video Pose Transformerscited by 5
  26. 2021PoseKernelLifter: Metric Lifting of 3D Human Pose using Soundcited by 4
  27. 2023GATOR: Graph-Aware Transformer with Motion-Disentangled Regression for Human Mesh Recovery from a 2D Posecited by 4
  28. 2023MPM: A Unified 2D-3D Human Pose Representation via Masked Pose Modelingcited by 4
  29. 2023Improving the Robustness of 3D Human Pose Estimation: A Benchmark and Learning from Noisy Inputcited by 4
  30. 2024EgoPoseFormer: A Simple Baseline for Stereo Egocentric 3D Human Pose Estimationcited by 4
  31. 2024CHAMP: Conformalized 3D Human Multi-Hypothesis Pose Estimatorscited by 4
  32. 2025HGMamba: Enhancing 3D Human Pose Estimation with a HyperGCN-Mamba Networkcited by 4
  33. 2022Adaptive Split-Fusion Transformercited by 3
  34. 2023Human Pose Estimation from Ambiguous Pressure Recordings with Spatio-temporal Masked Transformerscited by 3
  35. 2023Occlusion Robust 3D Human Pose Estimation with StridedPoseGraphFormer and Data Augmentationcited by 3
  36. 2024Your Turn: At Home Turning Angle Estimation for Parkinson's Disease Severity Assessmentcited by 3
  37. 2024Enhancing 3D Human Pose Estimation Amidst Severe Occlusion with Dual Transformer Fusioncited by 3
  38. 2024Optimizing Local-Global Dependencies for Accurate 3D Human Pose Estimationcited by 3

50 most cited of 94 All 94 (JSON)