Abstract: This paper aims at high-accuracy 3D object detection in autonomous driving scenario. We propose Multi-View 3D networks (MV3D), a sensory-fusion framework that takes both LIDAR point cloud ...
Abstract: Automated 3D reconstruction of faces from images is challenging if the image material is difficult in terms of pose, lighting, occlusions and facial expressions, and if the initial 2D ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results