Heterogeneous Multi-task Learning for Human Pose Estimation with Deep Convolutional Neural Network
This work addresses human pose estimation for computer vision applications, but it is incremental as it builds on existing multi-task learning and deep network approaches.
The authors tackled human pose estimation from monocular images by proposing a heterogeneous multi-task learning framework that simultaneously learns pose-joint regression and body-part detection in a deep convolutional neural network, resulting in competitive and state-of-the-art results on several datasets.
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.