Drive&Segment: Unsupervised Semantic Segmentation of Urban Scenes via Cross-modal Distillation

TL;DR

This paper introduces a novel unsupervised method for semantic segmentation of urban scenes using synchronized LiDAR and camera data, leveraging cross-modal distillation and object proposals to train models without manual annotations.

Contribution

The work presents a new cross-modal unsupervised learning approach that uses LiDAR-based object proposals and distillation to train semantic segmentation models without manual labels.

Findings

Achieves significant improvements over state-of-the-art on multiple datasets.

Demonstrates generalization without fine-tuning across diverse urban scenes.

Utilizes a transformer-based model trained with pseudo-classes from LiDAR data.

Abstract

This work investigates learning pixel-wise semantic image segmentation in urban scenes without any manual annotation, just from the raw non-curated data collected by cars which, equipped with cameras and LiDAR sensors, drive around a city. Our contributions are threefold. First, we propose a novel method for cross-modal unsupervised learning of semantic image segmentation by leveraging synchronized LiDAR and image data. The key ingredient of our method is the use of an object proposal module that analyzes the LiDAR point cloud to obtain proposals for spatially consistent objects. Second, we show that these 3D object proposals can be aligned with the input images and reliably clustered into semantically meaningful pseudo-classes. Finally, we develop a cross-modal distillation approach that leverages image data partially annotated with the resulting pseudo-classes to train a transformer-based model for image semantic segmentation. We show the generalization capabilities of our method by testing on four different testing datasets (Cityscapes, Dark Zurich, Nighttime Driving and ACDC) without any finetuning, and demonstrate significant improvements compared to the current state of the art on this problem. See project webpage https://vobecant.github.io/DriveAndSegment/ for the code and more.