Shape based detection at MC Lab

Shape based detection has advantages such as invariant to changes of appearance. However, it’s a very difficult task due to the problem of accidental alignment to background clutters. Only a small amount of literatures have addressed the problem. In this project, we aim to develop a framework based on holistic matching without breaking the model into parts and alleviating the problem of accidental alignment.

Paper anoymously submitted to CVPR2009.

Geolife at MSRA

GeoLife is a project that manages and understands the collected GPS data for geographical applications and provides ways to effectively visualize, organize, fast retrieve and understand GPS track logs for both personal and public use.

Papers on[ACM GIS'08, Ubicomp'08]

Contour Grouping at MC Lab

In the project of Contour grouping using Partial Similarity, a novel algorithm is introduced to group contours from clutter images by integrating high-level information (prior of part segments) and low-level information (edges of segmentations of clutter images). The partial shape similarity between these two levels of information is embedded into the particle filter framework, an effective recursively estimating model. The results illustrate quite good performance.

Paper on PSIVT'09