The 3rd International Workshop on Human Activity Understanding from 3D Data (HAU3D13) will be held on in conjunction with the IEEE Computer Vision and Pattern Recognition (CVPR) Conference, Portland, Oregon, June 23-28, 2013
Wokshop Date: Monday, June 24, 2013
Automatic analysis of human motion has been one of the most active research topics in computer vision due to the scientific challenges of the problem and the wide range of applications. Such applications include intelligent video surveillance; human-computer-interface (HCI); intelligent humanoid robots; diagnosis, assessment and treatment of musculoskeletal disorders; sports analysis; realistic synthesis and animation of human motion; and monitoring of elderly and disabled people at home. Extensive study has been conducted in the past decade using 2D visual information captured by single or multiple cameras. However, the problem, especially robust and viewpoint independent recognition of diverse human actions and activities in a real environment is far from being solved.
Recent advances in 3D depth cameras using structured light or time-of-flight sensors, 3D information recovery from 2D images/videos, and the availability of portable human motion capture devices have been nurturing a potential breakthrough solution to the problem of human activity recognition by using 3D data. The release of Microsoft’s Kinect Sensors and ASUS’s Xtion Pro Live Sensors including their Software Development Kit (SDK) have provided a commercially viable approach and hardware platform to capture 3D data in real-time.
This workshop will bring together leading researchers in the related fields of 3D data capture, representation and invariant description of 3D data, human motion modeling and applications of human motion analysis, in order to advocate and promote research in human activity recognition using 3D data. The workshop aims to provide an interactive forum for researchers to disseminate their most recent research results, rigorously and systematically discuss challenges and potential solutions towards a robust human activity recognition using 3D data, and to promote new collaborations amongst the researchers.