Metrics
3 Record Views
Conference proceeding
Shedding Light on Darkness: Enhancing Object Detection Robustness with Synthetic Perturbations for Real-world Challenges
2023 IEEE Conference on Artificial Intelligence (CAI), pp.36-37
IEEE Conference on Artificial Intelligence (CAI) (Santa Clara, California, USA, 06/05/2023–06/06/2023)
08/2023
Web of Science ID: WOS:001046447800013
Abstract
Robustness against distribution shifts is crucial for object detection models in real-world applications. In this study, we evaluate the performance of four state-of-the-art models against natural perturbations, retrain them with synthetic perturbations using the AugLy augmentation package, and assess their improved performance against natural perturbations. Our empirical ablation study focuses on the brightness perturbation modality using the COCO 2017 and ExDARK datasets. Our findings suggest that synthetic perturbations can effectively improve model robustness against real-world distribution shifts, providing valuable insights for deploying robust object detection models in real-world scenarios.
Related links
Details
- Title
- Shedding Light on Darkness: Enhancing Object Detection Robustness with Synthetic Perturbations for Real-world Challenges
- Publication Details
- 2023 IEEE Conference on Artificial Intelligence (CAI), pp.36-37
- Resource Type
- Conference proceeding
- Conference
- IEEE Conference on Artificial Intelligence (CAI) (Santa Clara, California, USA, 06/05/2023–06/06/2023)
- Publisher
- IEEE
- Number of pages
- 2
- Copyright
- © 2023, IEEE
- Identifiers
- WOS:001046447800013; 99381512041806600
- Academic Unit
- Intelligent Systems and Robotics; Hal Marcus College of Science and Engineering
- Language
- English