How Hard Is Snow? A Paired Domain Adaptation Dataset for Clear and Snowy Weather: CADC+

2025-06-19

AI Summary

Existing 3D detection evaluation under snow conditions is hindered by the scarcity of real, temporally and spatially aligned snowy–clear weather data pairs; mainstream datasets suffer from insufficient annotations or rely on synthetic desnowed images, introducing artifacts and domain shift. Method: We introduce CADC+, the first real-world, spatiotemporally aligned winter–summer domain adaptation dataset, capturing LiDAR point clouds under snowy and clear conditions on identical roads at matched times—eliminating non-snow-related domain shifts. We propose a pairing strategy grounded in spatiotemporal consistency and geometric registration, enabling the first artifact-free, real snowy–clear data pairing. Contribution/Results: We establish a robustness analysis framework that characterizes snow-induced hybrid uncertainty (aleatoric + epistemic). Evaluation on CADC+ reveals significant performance degradation across state-of-the-art 3D detectors in both localization and classification, establishing a reproducible benchmark for weather-robust perception.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

The impact of snowfall on 3D object detection performance remains underexplored. Conducting such an evaluation requires a dataset with sufficient labelled data from both weather conditions, ideally captured in the same driving environment. Current driving datasets with LiDAR point clouds either do not provide enough labelled data in both snowy and clear weather conditions, or rely on de-snowing methods to generate synthetic clear weather. Synthetic data often lacks realism and introduces an additional domain shift that confounds accurate evaluations. To address these challenges, we present CADC+, the first paired weather domain adaptation dataset for autonomous driving in winter conditions. CADC+ extends the Canadian Adverse Driving Conditions dataset (CADC) using clear weather data that was recorded on the same roads and in the same period as CADC. To create CADC+, we pair each CADC sequence with a clear weather sequence that matches the snowy sequence as closely as possible. CADC+ thus minimizes the domain shift resulting from factors unrelated to the presence of snow. We also present some preliminary results using CADC+ to evaluate the effect of snow on 3D object detection performance. We observe that snow introduces a combination of aleatoric and epistemic uncertainties, acting as both noise and a distinct data domain.

Problem

• Impact of snowfall on 3D object detection performance
• Lack of paired real-world snowy and clear weather datasets
• Domain shift issues in synthetic de-snowed data

Innovation

• Paired dataset for snowy and clear weather
• Minimizes domain shift with matched sequences
• Evaluates snow impact on 3D object detection