Panoptic-aware Image-to-Image Translation
This work addresses the challenge of generating high-fidelity translated images with accurate object details for applications like computer vision, though it is incremental by building on existing GAN and segmentation techniques.
The paper tackles the problem of low fidelity and poor object recognition in image-to-image translation for complex scenes by introducing PanopticGAN, which uses panoptic segmentation to align object content with style codes. The method achieved significant improvements in both image quality and object recognition performance compared to competing approaches.
Despite remarkable progress in image translation, the complex scene with multiple discrepant objects remains a challenging problem. The translated images have low fidelity and tiny objects in fewer details causing unsatisfactory performance in object recognition. Without thorough object perception (i.e., bounding boxes, categories, and masks) of images as prior knowledge, the style transformation of each object will be difficult to track in translation. We propose panoptic-aware generative adversarial networks (PanopticGAN) for image-to-image translation together with a compact panoptic segmentation dataset. The panoptic perception (i.e., foreground instances and background semantics of the image scene) is extracted to achieve alignment between object content codes of the input domain and panoptic-level style codes sampled from the target style space, then refined by a proposed feature masking module for sharping object boundaries. The image-level combination between content and sampled style codes is also merged for higher fidelity image generation. Our proposed method was systematically compared with different competing methods and obtained significant improvement in both image quality and object recognition performance.