TFPose: Direct Human Pose Estimation with Transformers

Abstract

We propose a human pose estimation framework that solves the task in the regression-based fashion. Unlike previous regression-based methods, which often fall behind those state-of-the-art methods, we formulate the pose estimation task into a sequence prediction problem that can effectively be solved by transformers. Our framework is simple and direct, bypassing the drawbacks of the heatmap-based pose estimation. Moreover, with the attention mechanism in transformers, our proposed framework is able to adaptively attend to the features most relevant to the target keypoints, which largely overcomes the feature misalignment issue of previous regression-based methods and considerably improves the performance. Importantly, our framework can inherently take advantages of the structured relationship between keypoints. Experiments on the MS-COCO and MPII datasets demonstrate that our method can significantly improve the state-of-the-art of regression-based pose estimation and perform comparably with the best heatmap-based pose estimation methods.

15 pages

References in corpus (23)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Deep Residual Learning for Image Recognition
• A Survey on Visual Transformer
• DeepPose: Human Pose Estimation via Deep Neural Networks
• Stacked Hourglass Networks for Human Pose Estimation
• Numerical Coordinate Regression with Convolutional Neural Networks
• Cascaded Pyramid Network for Multi-Person Pose Estimation
• Pre-Trained Image Processing Transformer
• Rethinking on Multi-Stage Networks for Human Pose Estimation
• Multi-Context Attention for Human Pose Estimation
• DirectPose: Direct End-to-End Multi-Person Pose Estimation
• HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation
• Towards Accurate Multi-person Pose Estimation in the Wild
• End-to-End Video Instance Segmentation with Transformers
• Human Pose Estimation with Spatial Contextual Information
• Compositional Human Pose Regression
• End-to-End Human Pose and Mesh Reconstruction with Transformers
• The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation
• MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers
• Human Pose Estimation using Deep Consensus Voting
• Rethinking the Heatmap Regression for Bottom-up Human Pose Estimation
• Integral Human Pose Regression

Cited by in corpus (8)

• A Survey on Visual Transformer
• Vision-Based Human Pose Estimation via Deep Learning: A Survey
• TransFusion: Cross-view Fusion with Transformer for 3D Human Pose Estimation
• SHaRPose: Sparse High-Resolution Representation for Human Pose Estimation
• Temporal Action Proposal Generation with Transformers
• Test-Time Personalization with a Transformer for Human Pose Estimation
• Boosting Salient Object Detection with Transformer-based Asymmetric Bilateral U-Net
• Semi-Supervised 3D Hand-Object Poses Estimation with Interactions in Time