3D Human Pose Estimation and Tracking across a Network of RGB and RGB-D Cameras

3D Human Pose Estimation and Tracking across a Network of RGB and RGB-D Cameras

Internship Description

In this project, we aim to construct a camera network of RGB and RGB-D sensors (Kinect) that can be efficiently streamed across the IP network. We will develop and implement robust algorithms (based on sparse and low-rank pose/appearance representations) that enable persistent tracking across the camera network. Moreover, by accumulating and transferring the RGB-D model of each moving individual from one camera view to another, 3D pose can be persistently estimated. Our proposed method will be applied to multi-human markerless augmented reality in the NexCave of the VCC center.‚Äč

Deliverables/Expectations

1. a hybrid network of RGB and RGB-D cameras setup in the NexCave to be used for AR and visualization purposes. 2. a paper submission to the European Conference on Computer Vision (ECCV 2014), which is a top-tier conference in the field 3. a large-scale dataset of multi-view RGB-D video feeds, where each individual is tracked over time and his/her pose is manually labeled. This dataset will be made publicly available to the research community for future algorithm evaluation and comparison

Faculty Name

Field of Study

Computer Vision, Machine Learning, Image Processing