R-CNNs for Pose Estimation and Action Detection

Abstract

We present convolutional neural networks for the tasks of keypoint (pose) prediction and action classification of people in unconstrained images. Our approach involves training an R-CNN detector with loss functions depending on the task being tackled. We evaluate our method on the challenging PASCAL VOC dataset and compare it to previous leading approaches. Our method gives state-of-the-art results for keypoint and action prediction. Additionally, we introduce a new dataset for action detection, the task of simultaneously localizing people and classifying their actions, and present results using our approach.

References in corpus (2)

• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• DeepPose: Human Pose Estimation via Deep Neural Networks

Cited by in corpus (24)

• Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods
• Dynamic Feature Integration for Simultaneous Detection of Salient Object, Edge and Skeleton
• Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
• Synthesizing Training Images for Boosting Human 3D Pose Estimation
• Deep Multi-Task Learning for Malware Image Classification
• 6D Pose Estimation with Combined Deep Learning and 3D Vision Techniques for a Fast and Accurate Object Grasping
• Video-based Human Action Recognition using Deep Learning: A Review
• End-to-End Multi-Task Learning with Attention
• Hypercolumns for Object Segmentation and Fine-grained Localization
• Group Ensemble: Learning an Ensemble of ConvNets in a single ConvNet
• Convolutional Models for Joint Object Categorization and Pose Estimation
• PaStaNet: Toward Human Activity Knowledge Engine
• Intention Recognition of Pedestrians and Cyclists by 2D Pose Estimation
• Single Image Action Recognition by Predicting Space-Time Saliency
• Cross-Domain Self-supervised Multi-task Feature Learning using Synthetic Imagery
• Autonomous Navigation in Dynamic Environments: Deep Learning-Based Approach
• MPM: Joint Representation of Motion and Position Map for Cell Tracking
• ACD: Action Concept Discovery from Image-Sentence Corpora
• Convolutional Channel Features
• Learning Action Concept Trees and Semantic Alignment Networks from Image-Description Data
• Understanding and Predicting The Attractiveness of Human Action Shot
• Robust Optimization for Deep Regression
• Nuisance-Label Supervision: Robustness Improvement by Free Labels
• A Multi-Modal Approach to Infer Image Affect