pith. sign in

arxiv: 2411.06632 · v1 · pith:KH65CARGnew · submitted 2024-11-10 · 💻 cs.CV · cs.LG· cs.RO

Few-shot Semantic Learning for Robust Multi-Biome 3D Semantic Mapping in Off-Road Environments

classification 💻 cs.CV cs.LGcs.RO
keywords semanticmiouoff-roadsegmentationapproachbiomesclassesconditions
0
0 comments X
read the original abstract

Off-road environments pose significant perception challenges for high-speed autonomous navigation due to unstructured terrain, degraded sensing conditions, and domain-shifts among biomes. Learning semantic information across these conditions and biomes can be challenging when a large amount of ground truth data is required. In this work, we propose an approach that leverages a pre-trained Vision Transformer (ViT) with fine-tuning on a small (<500 images), sparse and coarsely labeled (<30% pixels) multi-biome dataset to predict 2D semantic segmentation classes. These classes are fused over time via a novel range-based metric and aggregated into a 3D semantic voxel map. We demonstrate zero-shot out-of-biome 2D semantic segmentation on the Yamaha (52.9 mIoU) and Rellis (55.5 mIoU) datasets along with few-shot coarse sparse labeling with existing data for improved segmentation performance on Yamaha (66.6 mIoU) and Rellis (67.2 mIoU). We further illustrate the feasibility of using a voxel map with a range-based semantic fusion approach to handle common off-road hazards like pop-up hazards, overhangs, and water features.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Multistep Belief Space Dynamics Learning For Risk-Aware Control

    cs.RO 2026-05 unverdicted novelty 5.0

    A structured learning approach for multistep distributional dynamics in belief space enables real-time risk-aware MPC, validated via ablation on real off-road data and deployment on a full-sized vehicle.