LeAD-M3D: Leveraging Asymmetric Distillation for Real-time Monocular 3D Detection
This work addresses the challenge of achieving both high accuracy and real-time efficiency in monocular 3D detection for applications like autonomous driving, representing a significant advancement rather than an incremental improvement.
The paper tackled the problem of real-time monocular 3D object detection, which suffers from depth ambiguity and high computational costs, by introducing LeAD-M3D, a method that achieved state-of-the-art accuracy on datasets like KITTI and Waymo and ran up to 3.6x faster than prior high-accuracy methods without using extra modalities like LiDAR.
Real-time monocular 3D object detection remains challenging due to severe depth ambiguity, viewpoint shifts, and the high computational cost of 3D reasoning. Existing approaches either rely on LiDAR or geometric priors to compensate for missing depth, or sacrifice efficiency to achieve competitive accuracy. We introduce LeAD-M3D, a monocular 3D detector that achieves state-of-the-art accuracy and real-time inference without extra modalities. Our method is powered by three key components. Asymmetric Augmentation Denoising Distillation (A2D2) transfers geometric knowledge from a clean-image teacher to a mixup-noised student via a quality- and importance-weighted depth-feature loss, enabling stronger depth reasoning without LiDAR supervision. 3D-aware Consistent Matching (CM3D) improves prediction-to-ground truth assignment by integrating 3D MGIoU into the matching score, yielding more stable and precise supervision. Finally, Confidence-Gated 3D Inference (CGI3D) accelerates detection by restricting expensive 3D regression to top-confidence regions. Together, these components set a new Pareto frontier for monocular 3D detection: LeAD-M3D achieves state-of-the-art accuracy on KITTI and Waymo, and the best reported car AP on Rope3D, while running up to 3.6x faster than prior high-accuracy methods. Our results demonstrate that high fidelity and real-time efficiency in monocular 3D detection are simultaneously attainable - without LiDAR, stereo, or geometric assumptions.