PanDA: Panoptic Data Augmentation
This addresses the problem of efficient data augmentation for panoptic segmentation in computer vision, offering a novel approach that is incremental in method but broad in application.
The paper tackles the challenge of panoptic segmentation by introducing PanDA, a pixel-space data augmentation method that requires no extra data or training and is computationally cheap. It shows robust performance gains across models, backbones, and datasets, with concrete improvements in panoptic segmentation, instance segmentation, and detection.
The recently proposed panoptic segmentation task presents a significant challenge of image understanding with computer vision by unifying semantic segmentation and instance segmentation tasks. In this paper we present an efficient and novel panoptic data augmentation (PanDA) method which operates exclusively in pixel space, requires no additional data or training, and is computationally cheap to implement. By retraining original state-of-the-art models on PanDA augmented datasets generated with a single frozen set of parameters, we show robust performance gains in panoptic segmentation, instance segmentation, as well as detection across models, backbones, dataset domains, and scales. Finally, the effectiveness of unrealistic-looking training images synthesized by PanDA suggest that one should rethink the need for image realism for efficient data augmentation.