Lite-FPN for Keypoint-based Monocular 3D Object Detection

TL;DR

This paper introduces Lite-FPN, a lightweight feature pyramid network that enhances multi-scale detection in monocular 3D object detection, along with an attention loss to improve localization accuracy, resulting in better speed and precision.

Contribution

The paper proposes Lite-FPN for efficient multi-scale feature fusion and a novel attention loss to align confidence with localization quality in keypoint-based monocular 3D detection.

Findings

Significant accuracy improvements on KITTI dataset

Enhanced detection speed with Lite-FPN

Better alignment of confidence and localization with attention loss

Abstract

3D object detection with a single image is an essential and challenging task for autonomous driving. Recently, keypoint-based monocular 3D object detection has made tremendous progress and achieved great speed-accuracy trade-off. However, there still exists a huge gap with LIDAR-based methods in terms of accuracy. To improve their performance without sacrificing efficiency, we propose a sort of lightweight feature pyramid network called Lite-FPN to achieve multi-scale feature fusion in an effective and efficient way, which can boost the multi-scale detection capability of keypoint-based detectors. Besides, the misalignment between classification score and localization precision is further relieved by introducing a novel regression loss named attention loss. With the proposed loss, predictions with high confidence but poor localization are treated with more attention during the training phase. Comparative experiments based on several state-of-the-art keypoint-based detectors on the KITTI dataset show that our proposed methods manage to achieve significant improvements in both accuracy and frame rate. The code and pretrained models will be released at \url{https://github.com/yanglei18/Lite-FPN}.