Research map / Image generation / Image restoration
Semantically Structured Image Compression via Irregular Group-Based Decoupling
Ruoyu Feng, Yixin Gao, Xin Jin, Runsen Feng, Zhibo Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use irregular group-mask partitioning instead of bounding-box partitioning (ablation) | Group-mask semantically structured codec · COCO 2017 RoI reconstruction and pose estimation | bitrate at equal quality | Improved |
| use group mask instead of bounding box partitioning (ablation) | Group-mask semantically structured codec · COCO 2017 RoI reconstruction and pose estimation | visual quality and downstream task performance | No clear effect |
| add group-independent transform (GI Swin-Block) (ablation) | Group-mask semantically structured codec · COCO 2017 RoI reconstruction (people) | PSNR of RoI | Improved |
| add group-independent transform (ablation) | Group-mask semantically structured codec · COCO 2017 pose estimation | keypoint accuracy | Improved |
| structure the bitstream by semantics with group masks | Neural codec with ChARM · Kodak entire image reconstruction | PSNR vs bitrate | No clear effect |
| selective decoding with group mask versus SSIC and ELIC | Instance segmentation on COCO · COCO 2017 val, below 0.6 bpp | instance segmentation accuracy | Improved |
Closest papers by meaning
- Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion (ICML 2025 · similarity 0.83)
- Towards Unified Human Perception and Machine Understanding: Token Flow Guided Compression Framework (CVPR 2026 · similarity 0.83)
- Towards image compression with perfect realism at ultra-low bitrates (ICLR 2024 · similarity 0.82)
- Generative Latent Coding for Ultra-Low Bitrate Image Compression (CVPR 2024 · similarity 0.82)
- Switchable Token-Specific Codebook Quantization For Face Image Compression (NeurIPS 2025 · similarity 0.81)
- Once-for-All: Controllable Generative Image Compression with Dynamic Granularity Adaptation (ICLR 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
