Computer Vision News - December 2021

10 Congrats, Doctor! Muxingzi Li recently completed her Ph.D. at Inria Sophia Antipolis-Méditerranée. Her research interest lies at the intersection of geometry processing, computer vision and graphics, with the goal of enabling compact representations of objects from low-budget image data. She received her B.A. degree from University of Oxford and her MSc degree from KAUST. She has also spent time at Alibaba DAMO Academy. She will start as a researcher at miHoYo. Congrats, Doctor Muxingzi! Geometric approximation of urban objects with compact and accurate representation is a challenging problem that concerns both computer vision and computer graphics communities. Existing literature mainly focuses on reconstruction from high-quality point clouds obtained by laser scanning which are too costly for many practical scenarios. This motivates the investigation into the problem of geometric approximation from low-budget image data. Dense reconstruction from a collection of images is made possible by recent advances in multi-view stereo techniques, yet the resulting point cloud is often far from perfect for generating a compact model. In particular, our goal is to describe the captured scenewitha compact and accurate representation. We propose two generic algorithms which address different aspects of image-based geometric approximation. The proposed algorithms could be used sequentially to form a pipeline for geometric approximation of an urban object from a set of image data, consisting of an overhead shot for coarse model extraction and multi-view stereo data for point cloud generation.