Section: New Results

Object Detection from RGB-Depth images

Paticipant: Siddhartha Chandra, Iasonas Kokkinos

In [11] we explore RGB-Depth representations for the training of Deformable Models. and describe strategies to improve an object detection pipeline by introducing viewpoint based mixture components. Our contributions are threefold. First, we use surface-based object representations (3D mesh models) from available 3D object model repositories to learn strongly supervised viewpoint classifiers. Second, we develop a geometric dataset augmentation scheme that uses scene geometry to ‘take another look’ at the training data, simulating the effect of camera viewpoint changes. Third, to better exploit depth information, we develop a novel depth-based dense feature extraction method that provides a robust statistical description of scene geometry. We evaluate our learned detectors on the common NYU dataset, and demonstrate that each of our advances results in systematic performance improvements over the traditional detection pipeline.