1. [Publications](/publications) 2. Hand Pose Estimation via Latent 2.5 D Heatmap Regression # Hand Pose Estimation via Latent 2.5 D Heatmap Regression ![Publication image](/sites/default/files/styles/wide/public/default_images/default.jpeg?itok=qUFsuJCP "Publication image") Estimating the 3D pose of a hand is an essential part of human-computer interaction. Estimating 3D pose using depth or multi- view sensors has become easier with recent advances in computer vision, however, regressing pose from a single RGB image is much less straight- forward. The main difficulty arises from the fact that 3D pose requires some form of depth estimates, which are ambiguous given only an RGB image. In this paper we propose a new method for 3D hand pose estima- tion from a monocular image through a novel 2.5D pose representation. Our new representation estimates pose up to a scaling factor, which can be estimated additionally if a prior of the hand size is given. We im- plicitly learn depth maps and heatmap distributions with a novel CNN architecture. Our system achieves state-of-the-art accuracy for 2D and 3D hand pose estimation on several challenging datasets in presence of severe occlusions. ## Authors [Umar Iqbal](/person/umar-iqbal) [Pavlo Molchanov](/person/pavlo-molchanov) [Thomas Breuel](/person/thomas-breuel) Juergen Gall (University of Bonn, Germany) [Jan Kautz](/person/jan-kautz) ## Publication Date Thursday, September 13, 2018 ## Published in ECCV2018 ## Research Area [Computer Vision](/research-area/computer-vision) [Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence) ## External Links [Paper PDF](https://arxiv.org/pdf/1804.09534.pdf) ## Copyright This material is posted here with permission of the IEEE. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to .