Lite-FPN for Keypoint-based Monocular 3D Object Detection
This work addresses the problem of improving 3D object detection accuracy for autonomous driving without sacrificing efficiency, representing an incremental advancement in keypoint-based methods.
The paper tackles the accuracy gap between keypoint-based monocular 3D object detection and LIDAR-based methods by proposing Lite-FPN for efficient multi-scale feature fusion and an attention loss to align classification and localization, achieving significant improvements in accuracy and frame rate on the KITTI dataset.
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}.