mlsanityData to 14 Oct 2026

Unsupervised Monocular Road Segmentation for Autonomous Driving via Scene Geometry

arXiv 2510.16790 ↗First posted 19 Oct 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper presents a fully unsupervised approach for binary road segmentation (road vs. non-road), eliminating the reliance on costly manually labeled datasets. The method leverages scene geometry and temporal cues to distinguish road from non-road regions. Weak labels are first generated from geometric priors, marking pixels above the horizon as non-road and a predefined quadrilateral in front of the vehicle as road. In a refinement stage, temporal consistency is enforced by tracking local feature points across frames and penalizing inconsistent label assignments using mutual information maximization. This enhances both precision and temporal stability. On the Cityscapes dataset, the model achieves an Intersection-over-Union (IoU) of 0.86, outperforming the competing unsupervised methods. These findings demonstrate the potential of combining geometric constraints and temporal consistency for scalable unsupervised road segmentation in autonomous driving.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 38 references that are in the map. First posted on arXiv 19 Oct 2025.

Builds on

29 of 38 references.

References · 29

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  4. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  5. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  6. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,241
  7. 2019Searching for MobileNetV3cited by 1,994
  8. 2018Invariant Information Clustering for Unsupervised Image Classification and Segmentationcited by 455
  9. 2022Unsupervised Semantic Segmentation by Distilling Feature Correspondencescited by 241
  10. 2021Unsupervised Semantic Segmentation by Contrasting Object Mask Proposalscited by 190
  11. 2022Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cutcited by 167
  12. 2022Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localizationcited by 150
17 more
  1. 2020Autoregressive Unsupervised Image Segmentationcited by 60
  2. 2022ACSeg: Adaptive Conceptualization for Unsupervised Semantic Segmentationcited by 42
  3. 2023Leveraging Hidden Positives for Unsupervised Semantic Segmentationcited by 40
  4. 2021TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentationcited by 38
  5. 2022Hierarchical Feature Alignment Network for Unsupervised Video Object Segmentationcited by 37
  6. 2022Unsupervised Video Object Segmentation via Prototype Memory Networkcited by 30
  7. 2022DeepCut: Unsupervised Segmentation using Graph Neural Networks Clusteringcited by 25
  8. 2022Drive&Segment: Unsupervised Semantic Segmentation of Urban Scenes via Cross-modal Distillationcited by 19
  9. 2023Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Samplingcited by 19
  10. 2021InfoSeg: Unsupervised Semantic Image Segmentation with Mutual Information Maximizationcited by 11
  11. 2021Deep Superpixel Cut for Unsupervised Image Segmentationcited by 6
  12. 2023Weakly Supervised Semantic Segmentation for Driving Scenescited by 6
  13. 2025Learning Motion and Temporal Cues for Unsupervised Video Object Segmentationcited by 4
  14. 2017Distantly Supervised Road Segmentationcited by 3
  15. 2017Minimizing Supervision for Free-space Segmentationcited by 3
  16. 2023FODVid: Flow-guided Object Discovery in Videoscited by 3
  17. 2023Expand-and-Quantize: Unsupervised Semantic Segmentation Using High-Dimensional Space and Product Quantizationcited by 3

Built on by

Not cited yet.