SpectraDINO adapts frozen DINOv2 backbones to multispectral data via per-modality adapters and staged distillation with cosine, contrastive, patch, and neighborhood-structure losses, achieving SOTA on object detection and segmentation benchmarks.
In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Person2Drive is a new benchmark that generates personalized driving datasets via simulation, quantifies styles with MMD and KL metrics, and adapts E2E-AD models using a style reward framework.
citing papers explorer
-
SpectraDINO: Bridging the Spectral Gap in Vision Foundation Models via Lightweight Adapters
SpectraDINO adapts frozen DINOv2 backbones to multispectral data via per-modality adapters and staged distillation with cosine, contrastive, patch, and neighborhood-structure losses, achieving SOTA on object detection and segmentation benchmarks.
-
Driving with A Thousand Faces: A Benchmark for Closed-Loop Personalized End-to-End Autonomous Driving
Person2Drive is a new benchmark that generates personalized driving datasets via simulation, quantifies styles with MMD and KL metrics, and adapts E2E-AD models using a style reward framework.