2021/04/18 by Zengyi Qin, Jinglu Wang, Qin, Zengyi +3 · 4 citations
Computer Science · Engineering · Mathematics · #Advanced Image and Video Retrieval Techniques #Advanced Neural Network Applications #Algorithm #Amodal perception #Artificial intelligence #Bounding overwatch #Computer science #Computer vision #Dimension (graph theory) #Image (mathematics) #Mathematics #Minimum bounding box #Monocular #Object (grammar) #Object detection #Pattern recognition (psychology) #Perception #Projection (relational algebra) #Robotics and Sensor-Based Localization #Task (project management) #cs.CV
paper · pdf · doi:10.48550/arxiv.2104.08797
published in arXiv (Cornell University) (Cornell University) · The IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
arxiv created 2021/04/18 · openalex publication_date 2021/04/18 · arxiv updated 2021/04/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Detecting and localizing objects in the real 3D space, which plays a crucial role in scene understanding, is particularly challenging given only a monocular image due to the geometric information loss during imagery projection. We propose MonoGRNet for the amodal 3D object detection from a monocular image via geometric reasoning in both the observed 2D projection and the unobserved depth dimension. MonoGRNet decomposes the monocular 3D object detection task into four sub-tasks including 2D object detection, instance-level depth estimation, projected 3D center estimation and local corner regression. The task decomposition significantly facilitates the monocular 3D object detection, allowing the target 3D bounding boxes to be efficiently predicted in a single forward pass, without using object proposals, post-processing or the computationally expensive pixel-level depth estimation utilized by previous methods. In addition, MonoGRNet flexibly adapts to both fully and weakly supervised learning, which improves the feasibility of our framework in diverse settings. Experiments are conducted on KITTI, Cityscapes and MS COCO datasets. Results demonstrate the promising performance of our framework in various scenarios.