Unsupervised Object Detection with LiDAR Clues

2020/11/25 by Hao Tian, Tian, Hao, Yuntao Chen +7 · 1 citation
Computer Science · Engineering · Mathematics · #Advanced Image and Video Retrieval Techniques #Advanced Neural Network Applications #Artificial intelligence #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Computer vision #Exploit #FOS: Computer and information sciences #Geography #Lidar #Mathematics #Object (grammar) #Object detection #Pattern recognition (psychology) #Point (geometry) #Point cloud #Process (computing) #Remote sensing #Robotics and Sensor-Based Localization #Set (abstract data type) #cs.CV

paper · pdf · doi:10.48550/arxiv.2011.12953

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2020/11/25 · arxiv created 2021/04/19 · arxiv updated 2021/04/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

Despite the importance of unsupervised object detection, to the best of our knowledge, there is no previous work addressing this problem. One main issue, widely known to the community, is that object boundaries derived only from 2D image appearance are ambiguous and unreliable. To address this, we exploit LiDAR clues to aid unsupervised object detection. By exploiting the 3D scene structure, the issue of localization can be considerably mitigated. We further identify another major issue, seldom noticed by the community, that the long-tailed and open-ended (sub-)category distribution should be accommodated. In this paper, we present the first practical method for unsupervised object detection with the aid of LiDAR clues. In our approach, candidate object segments based on 3D point clouds are firstly generated. Then, an iterative segment labeling process is conducted to assign segment labels and to train a segment labeling network, which is based on features from both 2D images and 3D point clouds. The labeling process is carefully designed so as to mitigate the issue of long-tailed and open-ended distribution. The final segment labels are set as pseudo annotations for object detection network training. Extensive experiments on the large-scale Waymo Open dataset suggest that the derived unsupervised object detection method achieves reasonable accuracy compared with that of strong supervision within the LiDAR visible range. Code shall be released.

Citations

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal\n Networks
• GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose
• Soft Sampling for Robust Object Detection
• HDNET: Exploiting HD Maps for 3D Object Detection
• IPOD: Intensive Point-based Object Detector for Point Cloud
• Frustum ConvNet: Sliding Frustums to Aggregate Local Point-Wise Features for Amodal 3D Object Detection
• Contrastive Multiview Coding
• Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks
• Self-labelling via simultaneous clustering and representation learning
• Scalability in Perception for Autonomous Driving: Waymo Open Dataset
• Deep Learning for 3D Point Clouds: A Survey
• Proposal Learning for Semi-Supervised Object Detection
• A Simple Framework for Contrastive Learning of Visual Representations
• Improved Baselines with Momentum Contrastive Learning
• Momentum Contrast for Unsupervised Visual Representation Learning
• Learning Multiple Layers of Features from Tiny Images

Related

• Vision-Language Guidance for LiDAR-based Unsupervised 3D Object Detection
• Weakly Supervised 3D Object Detection from Lidar Point Cloud
• Once Detected, Never Lost: Surpassing Human Performance in Offline LiDAR based 3D Object Detection
• Towards Unsupervised Object Detection From LiDAR Point Clouds
• HD-OOD3D: Supervised and Unsupervised Out-of-Distribution object detection in LiDAR data