Some Past Research Activities:
3D Shape Descriptors: research on 3D models compact representations that have particular properties (e.g. reliability, invariance to translation, rotation and scale, noise resistance … ) and that can be used to compare 3D shapes in a database measuring similarity.
Skeletonization Methods: algorithms to extract a shape feature that represents the general form of a 2D/3D object.
Mobile 3D Metric Reconstruction: development of an algorithm for fast metric reconstruction on mobile devices using a combination of image and inertial acceleration data.
Camera Pose Estimation: 3D positioning and orientation of an image taken in an outdoor environment using image-based retrieval techniques and Structure From Motion data.
Semantic Image Segmentation: semantic segmentation of an image via label transfer from an already labeled image set, taking advantage of sparse 3D structure to infer the category of superpixel in the novel image.