# Unsupervised Object Detection with LiDAR Clues

arXiv: https://arxiv.org/abs/2011.12953 · HTML: https://mlsanity.com/paper/2011.12953/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Nov 2020.

## Built on by (most cited first, 22 of 22)

- [Localizing Objects with Self-Supervised Transformers and no Labels](https://mlsanity.com/paper/2109.14279/) (2021)
- [Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals](https://mlsanity.com/paper/2102.06191/) (2021)
- [SAVi++: Towards End-to-End Object-Centric Learning from Real-World Videos](https://mlsanity.com/paper/2206.07764/) (2022)
- [Motion Inspired Unsupervised Perception and Prediction in Autonomous Driving](https://mlsanity.com/paper/2210.08061/) (2022)
- [POP-3D: Open-Vocabulary 3D Occupancy Prediction from Images](https://mlsanity.com/paper/2401.09413/) (2024)
- [Learning to Detect Mobile Objects from LiDAR Scans Without Labels](https://mlsanity.com/paper/2203.15882/) (2022)
- [Unsupervised 3D Perception with 2D Vision-Language Distillation for Autonomous Driving](https://mlsanity.com/paper/2309.14491/) (2023)
- [Class-aware Sounding Objects Localization via Audiovisual Correspondence](https://mlsanity.com/paper/2112.11749/) (2021)
- [Drive&Segment: Unsupervised Semantic Segmentation of Urban Scenes via Cross-modal Distillation](https://mlsanity.com/paper/2203.11160/) (2022)
- [4D Unsupervised Object Discovery](https://mlsanity.com/paper/2210.04801/) (2022)
- [OccFeat: Self-supervised Occupancy Feature Prediction for Pretraining BEV Segmentation Networks](https://mlsanity.com/paper/2404.14027/) (2024)
- [Unsupervised Object Localization in the Era of Self-Supervised ViTs: A Survey](https://mlsanity.com/paper/2310.12904/) (2023)
- [Reward Finetuning for Faster and More Accurate Unsupervised Object Discovery](https://mlsanity.com/paper/2310.19080/) (2023)
- [Shelf-Supervised Cross-Modal Pre-Training for 3D Object Detection](https://mlsanity.com/paper/2406.10115/) (2024)
- [Approaching Outside: Scaling Unsupervised 3D Object Detection from 2D Scene](https://mlsanity.com/paper/2407.08569/) (2024)
- [LaneCorrect: Self-supervised Lane Detection](https://mlsanity.com/paper/2404.14671/) (2024)
- [Verifiable Obstacle Detection](https://mlsanity.com/paper/2208.14403/) (2022)
- [HUNTER: Unsupervised Human-centric 3D Detection via Transferring Knowledge from Synthetic Instances to Real Scenes](https://mlsanity.com/paper/2403.02769/) (2024)
- [MOD-UV: Learning Mobile Object Detectors from Unlabeled Videos](https://mlsanity.com/paper/2405.14841/) (2024)
- [You Only Label Once: 3D Box Adaptation from Point Cloud to Image via Semi-Supervised Learning](https://mlsanity.com/paper/2211.09302/) (2022)
- [RBF Weighted Hyper-Involution for RGB-D Object Detection](https://mlsanity.com/paper/2310.00342/) (2023)
- [Unsupervised Multi-agent and Single-agent Perception from Cooperative Views](https://mlsanity.com/paper/2604.05354/) (2026)

## Abstract (arXiv)

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.