RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving

TL;DR

This paper introduces RTM3D, a real-time monocular 3D detection framework for autonomous driving that predicts object keypoints and uses geometric relationships to accurately determine 3D properties from a single image.

Contribution

The paper presents the first real-time monocular 3D detection system that predicts nine perspective keypoints and uses geometric constraints, achieving state-of-the-art results without external supervision.

Findings

Achieves real-time detection speed on KITTI benchmark.

Outperforms previous methods in 3D detection accuracy.

Operates with a small, efficient architecture.

Abstract

In this work, we propose an efficient and accurate monocular 3D detection framework in single shot. Most successful 3D detectors take the projection constraint from the 3D bounding box to the 2D box as an important component. Four edges of a 2D box provide only four constraints and the performance deteriorates dramatically with the small error of the 2D detector. Different from these approaches, our method predicts the nine perspective keypoints of a 3D bounding box in image space, and then utilize the geometric relationship of 3D and 2D perspectives to recover the dimension, location, and orientation in 3D space. In this method, the properties of the object can be predicted stably even when the estimation of keypoints is very noisy, which enables us to obtain fast detection speed with a small architecture. Training our method only uses the 3D properties of the object without the need for external networks or supervision data. Our method is the first real-time system for monocular image 3D detection while achieves state-of-the-art performance on the KITTI benchmark. Code will be released at https://github.com/Banconxuan/RTM3D.