2020/05/31 by Rui Fan, Hengli Wang, Bohuan Xue +4 · 1 citation
Computer Science · Engineering · Mathematics · #Advanced Vision and Imaging #Algorithm #Artificial intelligence #Combinatorics #Computer science #Computer vision #Conjecture #Estimator #Filter (signal processing) #Geometry #Ground truth #Image (mathematics) #Inverse #Mathematics #Optical measurement and interference techniques #Pixel #Robotics and Sensor-Based Localization #Statistics #Surface (topology) #cs.CV #cs.RO
paper · pdf · doi:10.1109/lra.2021.3067308
webpage: sites.google.com/view/3f2n, accepted to IEEE RA-L and ICRA'21
arxiv created 2021/03/08 · openalex publication_date 2021/03/18 · arxiv updated 2021/04/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
This letter proposes three-filters-to-normal (3F2N), an accurate and ultrafast surface normal estimator (SNE), which is designed for structured range sensor data, e.g., depth/disparity images. 3F2N SNE computes surface normals by simply performing three filtering operations (two image gradient filters in horizontal and vertical directions, respectively, and a mean/median filter) on an inverse depth image or a disparity image. Despite the simplicity of 3F2N SNE, no similar method already exists in the literature. To evaluate the performance of our proposed SNE, we created three large-scale synthetic datasets (easy, medium and hard) using 24 3D mesh models, each of which is used to generate 1800-2500 pairs of depth images (resolution: 480 × 640 pixels) and the corresponding ground-truth surface normal maps from different views. 3F2N SNE demonstrates the state-of-the-art performance, outperforming all other existing geometry-based SNEs, where the average angular errors with respect to the easy, medium and hard datasets are 1.66°, 5.69°and 15.31°, respectively. Furthermore, our C++ and CUDA implementations achieve a processing speed of over 260 Hz and 21 kHz, respectively. Our datasets and source code are publicly available at sites.google.com/view/3f2n.