Spatiotemporal Contrastive Learning for Cross-View Video Localization in Unstructured Off-road Terrains

TL;DR

This paper introduces MoViX, a self-supervised framework for cross-view video localization in challenging off-road environments, achieving high accuracy despite seasonal and perceptual ambiguities.

Contribution

MoViX is the first to learn viewpoint- and season-invariant representations for off-road localization using a self-supervised approach with novel sampling and aggregation strategies.

Findings

Achieves 93% within 25 meters accuracy on TartanDrive 2.0

Operates effectively with limited training data (under 30 minutes)

Generalizes well to new environments and different robot platforms

Abstract

Robust cross-view 3-DoF localization in GPS-denied, off-road environments remains challenging due to (1) perceptual ambiguities from repetitive vegetation and unstructured terrain, and (2) seasonal shifts that significantly alter scene appearance, hindering alignment with outdated satellite imagery. To address this, we introduce MoViX, a self-supervised cross-view video localization framework that learns viewpoint- and season-invariant representations while preserving directional awareness essential for accurate localization. MoViX employs a pose-dependent positive sampling strategy to enhance directional discrimination and temporally aligned hard negative mining to discourage shortcut learning from seasonal cues. A motion-informed frame sampler selects spatially diverse frames, and a lightweight temporal aggregator emphasizes geometrically aligned observations while downweighting ambiguous ones. At inference, MoViX runs within a Monte Carlo Localization framework, using a learned cross-view matching module in place of handcrafted models. Entropy-guided temperature scaling enables robust multi-hypothesis tracking and confident convergence under visual ambiguity. We evaluate MoViX on the TartanDrive 2.0 dataset, training on under 30 minutes of data and testing over 12.29 km. Despite outdated satellite imagery, MoViX localizes within 25 meters of ground truth 93% of the time, and within 50 meters 100% of the time in unseen regions, outperforming state-of-the-art baselines without environment-specific tuning. We further demonstrate generalization on a real-world off-road dataset from a geographically distinct site with a different robot platform.