mlsanityData to 7 Oct 2026

Enhanced 3D Human Pose Estimation from Videos by using Attention-Based Neural Network with Dilated Convolutions

arXiv 2103.03170 ↗First posted 4 Mar 2021cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

The attention mechanism provides a sequential prediction framework for learning spatial models with enhanced implicit temporal consistency. In this work, we show a systematic design (from 2D to 3D) for how conventional networks and other forms of constraints can be incorporated into the attention framework for learning long-range dependencies for the task of pose estimation. The contribution of this paper is to provide a systematic approach for designing and training of attention-based models for the end-to-end pose estimation, with the flexibility and scalability of arbitrary video sequences as input. We achieve this by adapting temporal receptive field via a multi-scale structure of dilated convolutions. Besides, the proposed architecture can be easily adapted to a causal model enabling real-time performance. Any off-the-shelf 2D pose estimation systems, e.g. Mocap libraries, can be easily integrated in an ad-hoc fashion. Our method achieves the state-of-the-art performance and outperforms existing methods by reducing the mean per joint position error to 33.4 mm on Human3.6M dataset.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 38 of its 62 references that are in the map. First posted on arXiv 4 Mar 2021.

Builds on

38 of 62 references.

References · 38

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  3. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,530
  4. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  5. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,447
  6. 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,421
  7. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  8. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  9. 2019On the Variance of the Adaptive Learning Rate and Beyondcited by 763
  10. 2016Keep it SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Imagecited by 692
  11. 2016Language Modeling with Gated Convolutional Networkscited by 688
  12. 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
26 more
  1. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
  2. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 401
  3. 2015Attention-Based Models for Speech Recognitioncited by 400
  4. 2017Learning from Synthetic Humanscited by 358
  5. 2019Lookahead Optimizer: k steps forward, 1 step backcited by 333
  6. 2013DeepPose: Human Pose Estimation via Deep Neural Networkscited by 305
  7. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 220
  8. 2019Semantic Graph Convolutional Networks for 3D Human Pose Regressioncited by 204
  9. 20163D Human Pose Estimation = 2D Pose Estimation + Matchingcited by 112
  10. 20182D/3D Pose Estimation and Action Recognition using Multitask Deep Learningcited by 103
  11. 2017Exploiting temporal information for 3D pose estimationcited by 89
  12. 20183D Human Pose Estimation in the Wild by Adversarial Learningcited by 86
  13. 2018Learning Monocular 3D Human Pose Estimation from Multi-view Imagescited by 70
  14. 2017Multi-Context Attention for Human Pose Estimationcited by 66
  15. 2016Neural Machine Translation in Linear Timecited by 65
  16. 2015Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Videocited by 54
  17. 2016Synthesizing Training Images for Boosting Human 3D Pose Estimationcited by 52
  18. 2016Deep Kinematic Pose Regressioncited by 40
  19. 2015Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimationcited by 37
  20. 2015ABCNN: Attention-Based Convolutional Neural Network for Modeling Sentence Pairscited by 35
  21. 20163D Human Pose Estimation Using Convolutional Neural Networks with 2D Pose Informationcited by 28
  22. 2015Direct Prediction of 3D Body Poses from Motion Compensated Sequencescited by 26
  23. 2014Heterogeneous Multi-task Learning for Human Pose Estimation with Deep Convolutional Neural Networkcited by 11
  24. 2020Anatomy-aware 3D Human Pose Estimation with Bone-based Pose Decompositioncited by 10
  25. 2017Adversarial Learning of Structure-Aware Fully Convolutional Networks for Landmark Localizationcited by 3
  26. 2020A Graph Attention Spatio-temporal Convolutional Network for 3D Human Pose Estimation in Videocited by 3

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 1; 2023: 2; 2024: 1; 2025: 1.

6 papers.

  1. 2021Time-Ordered Recent Event (TORE) Volumes for Event Camerascited by 73
  2. 2023A Hybrid ANN-SNN Architecture for Low-Power and Low-Latency Visual Perceptioncited by 13
  3. 2023Towards Single Camera Human 3D-Kinematicscited by 8
  4. 2024RCR: Robust Crowd Reconstruction with Upright Space from a Single Large-scene Imagecited by 2
  5. 2022Human Pose Driven Object Effects Recommendationcited by 1
  6. 2025HGFreNet: Hop-hybrid GraphFomer for 3D Human Pose Estimation with Trajectory Consistency in Frequency Domaincited by 0