MonoGRNet: A General Framework for Monocular 3D Object Detection

TL;DR

MonoGRNet introduces a unified framework for monocular 3D object detection that decomposes the task into sub-tasks, enabling efficient and flexible detection without heavy computational costs, and demonstrates promising results on multiple datasets.

Contribution

It presents MonoGRNet, a novel geometric reasoning-based framework that efficiently predicts 3D bounding boxes from monocular images in a single pass, adaptable to supervised and weakly supervised learning.

Findings

Effective 3D detection on KITTI, Cityscapes, MS COCO

Single forward pass prediction without proposals

Flexible learning modes improve applicability

Abstract

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.