Image Codecs
Image codecs are methods for compressing and decompressing images, aiming to minimize data size while preserving image quality. Current research focuses on developing universal codecs applicable across various compression formats (JPEG, HEVC, AVIF, etc.), improving rate-distortion performance through techniques like neural networks (including transformers and diffusion models), and optimizing codecs for specific tasks such as machine vision or super-resolution. These advancements are significant because they improve efficiency in image storage and transmission, impacting fields ranging from medical imaging to entertainment and enabling new applications like real-time high-resolution image processing on resource-constrained devices.
Papers
GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian Splatting
Xinjie Zhang, Xingtong Ge, Tongda Xu, Dailan He, Yan Wang, Hongwei Qin, Guo Lu, Jing Geng, Jun Zhang
Content-aware Masked Image Modeling Transformer for Stereo Image Compression
Xinjie Zhang, Shenyuan Gao, Zhening Liu, Jiawei Shao, Xingtong Ge, Dailan He, Tongda Xu, Yan Wang, Jun Zhang