Robustness of Segment Anything Model (SAM) for Autonomous Driving in Adverse Weather Conditions

06/23/2023

∙

Segment Anything Model (SAM) has gained considerable interest in recent times for its remarkable performance and has emerged as a foundational model in computer vision. It has been integrated in diverse downstream tasks, showcasing its strong zero-shot transfer capabilities. Given its impressive performance, there is a strong desire to apply SAM in autonomous driving to improve the performance of vision tasks, particularly in challenging scenarios such as driving under adverse weather conditions. However, its robustness under adverse weather conditions remains uncertain. In this work, we investigate the application of SAM in autonomous driving and specifically explore its robustness under adverse weather conditions. Overall, this work aims to enhance understanding of SAM's robustness in challenging scenarios before integrating it into autonomous driving vision tasks, providing valuable insights for future applications.

READ FULL TEXT

Robustness of Segment Anything Model (SAM) for Autonomous Driving in Adverse Weather Conditions

Autonomous Driving in Adverse Weather Conditions: A Survey

MoWE: Mixture of Weather Experts for Multiple Adverse Weather Removal

In Rain or Shine: Understanding and Overcoming Dataset Bias for Improving Robustness Against Weather Corruptions for Autonomous Vehicles

Vision in adverse weather: Augmentation using CycleGANs with various object detectors for robust perception in autonomous racing

4DenoiseNet: Adverse Weather Denoising from Adjacent Point Clouds

Investigating Value of Curriculum Reinforcement Learning in Autonomous Driving Under Diverse Road and Weather Conditions

Doubly Contrastive End-to-End Semantic Segmentation for Autonomous Driving under Adverse Weather

Robustness of Segment Anything Model (SAM) for Autonomous Driving in Adverse Weather Conditions

Related Research

Autonomous Driving in Adverse Weather Conditions: A Survey

MoWE: Mixture of Weather Experts for Multiple Adverse Weather Removal

In Rain or Shine: Understanding and Overcoming Dataset Bias for Improving Robustness Against Weather Corruptions for Autonomous Vehicles

Vision in adverse weather: Augmentation using CycleGANs with various object detectors for robust perception in autonomous racing

4DenoiseNet: Adverse Weather Denoising from Adjacent Point Clouds

Investigating Value of Curriculum Reinforcement Learning in Autonomous Driving Under Diverse Road and Weather Conditions

Doubly Contrastive End-to-End Semantic Segmentation for Autonomous Driving under Adverse Weather