# Zero-shot 3D General Obstacle Detection via Multimodal Foundation Models and Geometry

arXiv: https://arxiv.org/abs/2408.12322 · HTML: https://mlsanity.com/paper/2408.12322/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 31 references that are in the map.
- First posted on arXiv 22 Aug 2024.

## Builds on (most cited first, 27 of 27)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Masked-attention Mask Transformer for Universal Image Segmentation](https://mlsanity.com/paper/2112.01527/) (2021)
- [BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers](https://mlsanity.com/paper/2203.17270/) (2022)
- [Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D](https://mlsanity.com/paper/2008.05711/) (2020)
- [Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks](https://mlsanity.com/paper/2401.14159/) (2024)
- [BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation](https://mlsanity.com/paper/2205.13542/) (2022)
- [Point Transformer V3: Simpler, Faster, Stronger](https://mlsanity.com/paper/2312.10035/) (2023)
- [BEVFusion: A Simple and Robust LiDAR-Camera Fusion Framework](https://mlsanity.com/paper/2205.13790/) (2022)
- [Towards Open World Object Detection](https://mlsanity.com/paper/2103.02603/) (2021)
- [OpenMask3D: Open-Vocabulary 3D Instance Segmentation](https://mlsanity.com/paper/2306.13631/) (2023)
- [FUTR3D: A Unified Sensor Fusion Framework for 3D Detection](https://mlsanity.com/paper/2203.10642/) (2022)
- [PLA: Language-Driven Open-Vocabulary 3D Scene Understanding](https://mlsanity.com/paper/2211.16312/) (2022)
- [SegmentMeIfYouCan: A Benchmark for Anomaly Segmentation](https://mlsanity.com/paper/2104.14812/) (2021)
- [Segment Any Point Cloud Sequences by Distilling Vision Foundation Models](https://mlsanity.com/paper/2306.09347/) (2023)
- [Lost and Found: Detecting Small Road Hazards for Self-Driving Vehicles](https://mlsanity.com/paper/1609.04653/) (2016)
- [Learning Open-World Object Proposals without Learning to Classify](https://mlsanity.com/paper/2108.06753/) (2021)
- [Contrastive Lift: 3D Object Instance Segmentation by Slow-Fast Contrastive Fusion](https://mlsanity.com/paper/2306.04633/) (2023)
- [Detecting Everything in the Open World: Towards Universal Object Detection](https://mlsanity.com/paper/2303.11749/) (2023)
- [CoDA: Collaborative Novel Box Discovery and Cross-modal Alignment for Open-vocabulary 3D Object Detection](https://mlsanity.com/paper/2310.02960/) (2023)
- [POP-3D: Open-Vocabulary 3D Occupancy Prediction from Images](https://mlsanity.com/paper/2401.09413/) (2024)
- [UC-OWOD: Unknown-Classified Open World Object Detection](https://mlsanity.com/paper/2207.11455/) (2022)
- [Unsupervised 3D Perception with 2D Vision-Language Distillation for Autonomous Driving](https://mlsanity.com/paper/2309.14491/) (2023)
- [aiMotive Dataset: A Multimodal Dataset for Robust Autonomous Driving with Long-Range Perception](https://mlsanity.com/paper/2211.09445/) (2022)
- [NVRadarNet: Real-Time Radar Obstacle and Free Space Detection for Autonomous Driving](https://mlsanity.com/paper/2209.14499/) (2022)
- [ElC-OIS: Ellipsoidal Clustering for Open-World Instance Segmentation on LiDAR Data](https://mlsanity.com/paper/2303.04351/) (2023)

## Abstract (arXiv)

Detecting general obstacles is critical for autonomous driving, especially in long-tail scenarios with rare or unseen objects. Existing methods rely on supervision or predefined categories, limiting generalization. We propose a training-free approach that combines multimodal foundation models with geometric reasoning for 3D obstacle detection. Our key idea is to detect obstacles as deviations from the road surface, segmented in 2D and localized in 3D via temporal LiDAR aggregation. The pipeline operates in a zero-shot manner without task-specific training. Experiments show accurate localization up to 100 meters and 10-25% recall gains from foundation model priors, while enabling scalable autolabeling.