On the development of a robust, fast and lightweight keypoint descriptor

At the core of a myriad of tasks such as object recognition, tridimensional reconstruction and alignment resides the critical problem of correspondence. Hence, devising descriptors, which identify the entities to be matched and that are able to correctly and reliably establish pairs of corresponding points is of central importance. We introduce three novel descriptors that efficiently combine appearance and geometrical shape information from RGB-D images, and are largely invariant to rotation, illumination changes and scale transformations. Results of several experiments described here demonstrate that as far as precision and robustness are concerned, our descriptors compare favorably to standard descriptors in the literature. In addition, they outperfom the state-of-the-art CSHOT, which, as well as our descriptors, combines texture and geometry. Also, we use these new descriptors to detect and recognize objects under different illumination conditions to provide semantic information in a mapping task and we apply our descriptors for registering multiple indoor textured depth maps, and demonstrate that they are robust and provide reliable results even for sparsely textured and poorly illuminated scenes. Experimental results show that our descriptors are superior to the others in processing time, memory consumption, recognition rate and alignment quality. Keywords-Descriptors; Keypoints; RGB-D Images.