Developing a reliable vision system is a fundamental challenge for robotic technologies e. Current solutions proposed to improve model robustness usually rely on generic data augmentation techniques or employ costly test-time adaptation methods.
In addition, most approaches focus on addressing a single vision task typically, image recognition utilising synthetic data. In this paper, we introduce Per-corruption Adaptation of Normalization statistics PAN to enhance the model robustness of vision systems.
Our approach entails three key components: i a corruption type identification module, ii dynamic adjustment of normalization layer statistics based on identified corruption type, and iii real-time update of these statistics according to input data.
PAN can integrate seamlessly with any convolutional model for enhanced accuracy in several robot vision tasks. In our experiments, PAN obtains robust performance improvement on challenging real-world corrupted image datasets e.
A reliable perception system is one of the key components of autonomous robotics, both for outdoor e. Advancements in deep learning technologies have led to the development of robust models for various robotic-related computer vision tasks, such as object recognition [ 1 , 2 ] , detection [ 3 ] and semantic segmentation [ 4 , 5 ].
