DinoRADE: Full Spectral Radar-Camera Fusion with Vision Foundation Model Features for Multi-class Object Detection in Adverse Weather
2026-04-09 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed DinoRADE, a detection system that uses Radar and camera data together to better spot small and vulnerable road users like pedestrians, especially in bad weather. Their system improves on previous methods by combining detailed radar signals with smart image features from a strong vision model called DINOv3. They tested DinoRADE on a tough dataset with different weather conditions and showed it detects five types of objects more accurately than past approaches. This work helps make self-driving cars safer in challenging environments.
FMCW Radarmulti-modal perceptionadverse weathervulnerable road usersdeformable cross-attentionDINOv3 Vision Foundation ModelK-Radar datasetautonomous drivingobject detection
Authors
Christof Leitgeb, Thomas Puchleitner, Max Peter Ronecker, Daniel Watzenig
Abstract
Reliable and weather-robust perception systems are essential for safe autonomous driving and typically employ multi-modal sensor configurations to achieve comprehensive environmental awareness. While recent automotive FMCW Radar-based approaches achieved remarkable performance on detection tasks in adverse weather conditions, they exhibited limitations in resolving fine-grained spatial details particularly critical for detecting smaller and vulnerable road users (VRUs). Furthermore, existing research has not adequately addressed VRU detection in adverse weather datasets such as K-Radar. We present DinoRADE, a Radar-centered detection pipeline that processes dense Radar tensors and aggregates vision features around transformed reference points in the camera perspective via deformable cross-attention. Vision features are provided by a DINOv3 Vision Foundation Model. We present a comprehensive performance evaluation on the K-Radar dataset in all weather conditions and are among the first to report detection performance individually for five object classes. Additionally, we compare our method with existing single-class detection approaches and outperform recent Radar-camera approaches by 12.1%. The code is available under https://github.com/chr-is-tof/RADE-Net.