Here is a detailed caption for the image: --- The poster titled "Generative Latent Coding for Ultra-Low Bitrate Image Compression" presents the research by Zhaoyang Jia and colleagues from the University of Science and Technology of China and Microsoft Research Asia. Displayed at the CVPR conference in Seattle, WA, from June 17-23, 2024, the poster outlines a novel image compression method aimed at achieving significant bitrate savings. **Introduction Section:** - Illustrates the core concept of Generative Latent Coding (GLC) for image compression. - Highlights comparisons in pixel space, human perception space, and generative latent space with sample images showing variation in texture realism and semantic consistency. - States that GLC achieves a reduction in bitrate by 45% while maintaining the same FID, compared to state-of-the-art methods. **Methods Section:** - Provides a detailed overview of the GLC framework and progressive learning methods. - Showcases the transform coding process in VQVAE latent space and employs a code-prediction-based supervision loss. - Features categorical hyper module to enhance the compression process. **Results Section:** - Contains a visual comparison of original and compressed images under the GLC method. - Displays rate-distortion curves highlighting the performance improvements in VVC H.266, VVC H.266 (2021), Non-Hier. GAN Loss (EQ1999), and Non-Hier. QIM Loss (EQ1999). By effectively demonstrating the benefits and methodologies of GLC, this poster offers significant insights into advancements in image compression technology. --- This caption captures the various aspects of the research, providing a comprehensive overview of the poster content. Text transcribed from the image: *** 1958 ty of Science and Technology of Introduction Generative Latent Coding for Ultra-Low Bitrate Image Compression Zhaoyang jia¹* Jiahao Li² 1University of Science and Technology of China 2Microsoft Research Asia Bin Li² Houqiang Li¹ Yan Lu² Compression Space Mapping Original Pixel-Space Compressed Results ⭑ Latent-Space Compressed Results Nearest Space of Code 2 Code +1 DISTS = 0.20 SNR=20.3 DISTS = 0.09 SNR 21.8 Code +2 Pixel Space Human Perception Space Original Image Generative Latent Space Latent Space Compression Pixel Space Compression Methods Analysis Transform Analysis Analysis Transform Transformy Pixel Loss Synthesis Transform Synthesis Transform Synthesis Transform Generative Latent Coding Training Stage I Training Stage II Overall Framework & Progressive Learning Original Training Stage III Text Sketch 0.029 bpp Ours 0.023 bpp Ours 0.036 bpp Lower Texture Realism? Semantic Consistency? Higher We propose Generative Latent Coding (GLC) for ultra-low bitrate image coding • Transform coding in VQVAE latent space Using categorical hyper module Code-prediction-based supervision GLC achieves 45% bits saving under same FID compared with previous SOTA This work was done when Zhaoyang Jia was an intern at MSRA ga 9s ha Code-Pred-based Latent Loss (Stage) Code-Pred-based Supervision Code-red-based Supervision MISE Hyper AE VO-E Codebook C Code-Pred-based Pixel Loss (Stage Codebock Aux Code Predictor hvao Transform Coding in Latent Space Other features: . Code-Pred-based Supervision Cress Entropy Code-Prediction-Based loss 220 Variable bitrate in a single model Restoration and style transfer applications * Results CVPR SEATTLE, WA JUNE 17-21, 2024 HAC 0.041 bpp MS-ILLM 0.037 bpp Original Ours 0.038 bpp WC/H.266 FCC (ICPR2021) CelebAHO Visual Comparison --EVC (ICLR2023) -- TCM (CVPR2023) MS-ILLM (ICML20231 HIFIC (NeuriP52020) CUC 2020 Text Sketch OCMW2023) --GLC (Proposed) Rate-Distortion Curves 029 411 188