DeepPose: Human Pose Estimation via Deep Neural Networks

• Published in: 2014 IEEE Conference on Computer Vision and Pattern Recognition pp. 1653 - 1660
• Main Authors: Toshev, Alexander, Szegedy, Christian
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01.06.2014
• Subjects: Benchmarks; cascades; Computational modeling; Computer vision; Conferences; deep learning; Detectors; Estimates; Estimation; Human; human pose estimation; Joints; Measurement; Neural networks; Pattern recognition; Regression; Titanium dioxide; Training; Vectors

We propose a method for human pose estimation based on Deep Neural Networks (DNNs). The pose estimation is formulated as a DNN-based regression problem towards body joints. We present a cascade of such DNN regres- sors which results in high precision pose estimates. The approach has the advantage of...