Papers/2611.17438
🧪 Test?View on arXiv

Real-Time Diffusion Relighting as a Neural Sensor: Closing the Low-Light Domain Gap for Robotic Perception

Author1, Author2, Author3, Author4, Author5

low-light enhancementrobotic perceptiongenerative models
2611.17438
Builder Relevance
80%
19h ago

Abstract

The paper presents a method to enhance robotic perception in low-light conditions using a real-time generative neural sensor.

Reality Card

Core Claim

The proposed method achieves a new state-of-the-art performance on the ExDark low-light benchmark with 87.1% mAP and a 100% lap completion rate in a closed-loop autonomous driving simulator.

Method / Result

87.1% mAP on the ExDark low-light benchmark.

Limitations

The reliance on a large-scale image-to-image diffusion model may pose challenges for reproducibility in different environments.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.
← Back to all papers