no code implementations • 7 Aug 2023 • Xiang Feng, Kaizhang Kang, Fan Pei, Huakeng Ding, Jinjiang You, Ping Tan, Kun Zhou, Hongzhi Wu
We propose a novel framework to automatically learn to aggregate and transform photometric measurements from multiple unstructured views into spatially distinctive and view-invariant low-level features, which are fed to a multi-view stereo method to enhance 3D reconstruction.