Heterogeneous Multi-task Learning for Human Pose Estimation with Deep Convolutional Neural Network

Abstract

We propose an heterogeneous multi-task learning framework for human pose estimation from monocular image with deep convolutional neural network. In particular, we simultaneously learn a pose-joint regressor and a sliding-window body-part detector in a deep network architecture. We show that including the body-part detection task helps to regularize the network, directing it to converge to a good solution. We report competitive and state-of-art results on several data sets. We also empirically show that the learned neurons in the middle layer of our network are tuned to localized body parts.

References in corpus (2)

• Improving neural networks by preventing co-adaptation of feature detectors
• DeepPose: Human Pose Estimation via Deep Neural Networks

Cited by in corpus (13)

• DenseBox: Unifying Landmark Localization with End to End Object Detection
• Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
• Synthesizing Training Images for Boosting Human 3D Pose Estimation
• Transfer Learning of Artist Group Factors to Musical Genre Classification
• Convolutional Models for Joint Object Categorization and Pose Estimation
• DeepSkeleton: Skeleton Map for 3D Human Pose Regression
• Ranking Micro-Influencers: a Novel Multi-Task Learning and Interpretable Framework
• Learning Boost by Exploiting the Auxiliary Task in Multi-task Domain
• HF-UNet: Learning Hierarchically Inter-Task Relevance in Multi-Task U-Net for Accurate Prostate Segmentation
• Multi-task Learning by Leveraging the Semantic Information
• Distill-2MD-MTL: Data Distillation based on Multi-Dataset Multi-Domain Multi-Task Frame Work to Solve Face Related Tasksks, Multi Task Learning, Semi-Supervised Learning
• Reduced Rank Multivariate Kernel Ridge Regression
• W-net: Simultaneous segmentation of multi-anatomical retinal structures using a multi-task deep neural network