# HyperMODEST: Self-Supervised 3D Object Detection with Confidence Score Filtering

arXiv: https://arxiv.org/abs/2304.14446 · HTML: https://mlsanity.com/paper/2304.14446/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 31 references that are in the map.
- First posted on arXiv 27 Apr 2023.

## Builds on (most cited first, 23 of 23)

- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Center-based 3D Object Detection and Tracking](https://mlsanity.com/paper/2006.11275/) (2020)
- [PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud](https://mlsanity.com/paper/1812.04244/) (2018)
- [Confidence Regularized Self-Training](https://mlsanity.com/paper/1908.09822/) (2019)
- [Joint 3D Proposal Generation and Object Detection from View Aggregation](https://mlsanity.com/paper/1712.02294/) (2017)
- [PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detection](https://mlsanity.com/paper/2102.00463/) (2021)
- [Group-Free 3D Object Detection via Transformers](https://mlsanity.com/paper/2104.00678/) (2021)
- [End-to-End Multi-View Fusion for 3D Object Detection in LiDAR Point Clouds](https://mlsanity.com/paper/1910.06528/) (2019)
- [ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection](https://mlsanity.com/paper/2103.05346/) (2021)
- [Train in Germany, Test in The USA: Making 3D Object Detectors Generalize](https://mlsanity.com/paper/2005.08139/) (2020)
- [xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentation](https://mlsanity.com/paper/1911.12676/) (2019)
- [Point Density-Aware Voxels for LiDAR 3D Object Detection](https://mlsanity.com/paper/2203.05662/) (2022)
- [Back-tracing Representative Points for Voting-based 3D Object Detection in Point Clouds](https://mlsanity.com/paper/2104.06114/) (2021)
- [AFDet: Anchor Free One Stage 3D Object Detection](https://mlsanity.com/paper/2006.12671/) (2020)
- [Learning to Detect Mobile Objects from LiDAR Scans Without Labels](https://mlsanity.com/paper/2203.15882/) (2022)
- [Deep Co-Training with Task Decomposition for Semi-Supervised Domain Adaptation](https://mlsanity.com/paper/2007.12684/) (2020)
- [InfoFocus: 3D Object Detection for Autonomous Driving with Dynamic Information Modeling](https://mlsanity.com/paper/2007.08556/) (2020)
- [Simultaneous Edge Alignment and Learning](https://mlsanity.com/paper/1808.01992/) (2018)
- [Exploiting Playbacks in Unsupervised Domain Adaptation for 3D Object Detection](https://mlsanity.com/paper/2103.14198/) (2021)
- [Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception](https://mlsanity.com/paper/2203.11405/) (2022)
- [Drive&Segment: Unsupervised Semantic Segmentation of Urban Scenes via Cross-modal Distillation](https://mlsanity.com/paper/2203.11160/) (2022)
- [Simultaneous Localization, Mapping, and Manipulation for Unsupervised Object Discovery](https://mlsanity.com/paper/1411.0802/) (2014)
- [Contrastive Learning for Self-Supervised Pre-Training of Point Cloud Segmentation Networks With Image Data](https://mlsanity.com/paper/2301.07283/) (2023)

## Built on by (most cited first, 3 of 3)

- [LISO: Lidar-only Self-Supervised 3D Object Detection](https://mlsanity.com/paper/2403.07071/) (2024)
- [HUNTER: Unsupervised Human-centric 3D Detection via Transferring Knowledge from Synthetic Instances to Real Scenes](https://mlsanity.com/paper/2403.02769/) (2024)
- [OWL: Unsupervised 3D Object Detection by Occupancy Guided Warm-up and Large Model Priors Reasoning](https://mlsanity.com/paper/2512.05698/) (2025)

## Abstract (arXiv)

Current LiDAR-based 3D object detectors for autonomous driving are almost entirely trained on human-annotated data collected in specific geographical domains with specific sensor setups, making it difficult to adapt to a different domain. MODEST is the first work to train 3D object detectors without any labels. Our work, HyperMODEST, proposes a universal method implemented on top of MODEST that can largely accelerate the self-training process and does not require tuning on a specific dataset. We filter intermediate pseudo-labels used for data augmentation with low confidence scores. On the nuScenes dataset, we observe a significant improvement of 1.6% in AP BEV in 0-80m range at IoU=0.25 and an improvement of 1.7% in AP BEV in 0-80m range at IoU=0.5 while only using one-fifth of the training time in the original approach by MODEST. On the Lyft dataset, we also observe an improvement over the baseline during the first round of iterative self-training. We explore the trade-off between high precision and high recall in the early stage of the self-training process by comparing our proposed method with two other score filtering methods: confidence score filtering for pseudo-labels with and without static label retention. The code and models of this work are available at https://github.com/TRAILab/HyperMODEST