Selected Projects

Improving Posture Classification Accuracy for Depth Sensor-based Human Activity Monitoring in Smart Environments

In this project, we propose a framework that accurately classifies the nature of the 3D postures obtained by Kinect using a max-margin classifier. Different from previous work in the area, we integrate the information about the reliability of the tracked joints in order to enhance the accuracy and robustness of our framework. We demonstrate how our framework can be applied to evaluate the user's posture and identify the postures that may result in musculoskeletal disorders. Such a system can be used in workplace such as offices and factories to reduce risk of injury.
Spatial Relationship Preserving Character Motion Adaptation

In this project, we propose a new method for editing and retargeting motions that involve close interactions between body parts of single or multiple articulated characters or between characters and a restricted environment. In such motions, the implicit spatial relationships between body parts/objects are important for capturing the scene semantics. We introduce a simple structure called an interaction mesh to represent such spatial relationships. We demonstrate its effectiveness and versatility in synthesizing a wide variety of motions with close interactions.
Synthesizing Close Human Interactions with Topology Constraints

In this project, we proposed new methods to analysize and generate character motions that invlove close contacts. We introduce the concept of topology coordinates, in which the topological relationships of the segments are embedded into the attributes. There is a wide range of applications of the new methods, for example, content-based retrieval and character animation synthesis.
Reactive Human Motions

In this project, we propose a new method to generate reactive motions for arbitrary external perturbations such as pushing, pulling, or hitting. The method can be applied to motions with arbitrary contact states such as standing, walking or running.