Monocular 3D Object Detection with Sequential Feature Association and Depth Hint Augmentation

Abstract

Monocular 3D object detection, with the aim of predicting the geometric properties of on-road objects, is a promising research topic for the intelligent perception systems of autonomous driving. Most state-of-the-art methods follow a keypoint-based paradigm, where the keypoints of objects are predicted and employed as the basis for regressing the other geometric properties. In this work, a unified network named as FADNet is presented to address the task of monocular 3D object detection. In contrast to previous keypoint-based methods, we propose to divide the output modalities into different groups according to the estimation difficulty of object properties. Different groups are treated differently and sequentially associated by a convolutional Gated Recurrent Unit. Another contribution of this work is the strategy of depth hint augmentation. To provide characterized depth patterns as hints for depth estimation, a dedicated depth hint module is designed to generate row-wise features named as depth hints, which are explicitly supervised in a bin-wise manner. The contributions of this work are validated by conducting experiments and ablation study on the KITTI benchmark. Without utilizing depth priors, post optimization, or other refinement modules, our network performs competitively against state-of-the-art methods while maintaining a decent running speed.

arXiv comment11 pages, 11 figures. This paper has been accepted by IEEE Transactions of Intelligent Vehicles. Digital Object Identifier: 10.1109/TIV.2022.3143954

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• YOLOv4: Optimal Speed and Accuracy of Object Detection
• YOLOv3: An Incremental Improvement
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
• Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex
• 3DSSD: Point-based 3D Single Stage Object Detector
• RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Driving
• Joint Monocular 3D Vehicle Detection and Tracking

Cited by in corpus (1)

• Surround-view Fisheye BEV-Perception for Valet Parking: Dataset, Baseline and Distortion-insensitive Multi-task Framework