Research map / Detection and segmentation / CNN features
Region-Adaptive Transform with Segmentation Prior for Image Compression
Yuxi Liu, Wenhan Yang, Huihui Bai, Yunchao Wei, Yao Zhao
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 |
|---|---|---|---|
| add Region-Adaptive Transform (RAT) to Zou2022 (ablation) | Zou2022 · CLIC Professional Valid | BD-rate (%): 0.0 -> -4.35 | Improved |
| add Scale Affine Layer (SAL) to Zou2022 (ablation) | Zou2022 · CLIC Professional Valid | BD-rate (%): 0.0 -> -2.93 | Improved |
| combine SAL and RAT (ablation) | Zou2022 · CLIC Professional Valid | BD-rate (%): 0.0 -> -8.50 | Improved |
| use class-agnostic masks as privileged information during training (ablation) | SegPIC · COCO-sample | BD-rate (%): +2.10 -> 0 (default) | Improved |
| use real masks instead of 4x4 grid at test time (ablation) | SegPIC · COCO-sample | BD-rate (%): 0 -> -0.11 | No clear effect |
| finer test grid (8x8) instead of 4x4 (ablation) | SegPIC · COCO-sample | BD-rate (%): 0 -> +0.46 | Hurt |
| add RAT and SAL to a small baseline without attention or autoregression (ablation) | BL-S · Kodak (anchor VTM-17.0) | BD-rate (%): +15.5 -> +3.87 | Improved |
| add RAT and SAL to Zou2022 | Zou2022 · Kodak encoding time | ms: 120 -> 142 | Hurt |
Closest papers by meaning
- Learned Image Compression via Sparse Attention and Adaptive Frequency (CVPR 2026 · similarity 0.87)
- REN: Fast and Efficient Region Encodings from Patch-Based Image Encoders (NeurIPS 2025 · similarity 0.86)
- R-MAE: Regions Meet Masked Autoencoders (ICLR 2024 · similarity 0.85)
- Few-Shot Domain Adaptation for Learned Image Compression (AAAI 2025 · similarity 0.85)
- Another Way to the Top: Exploit Contextual Clustering in Learned Image Coding (AAAI 2024 · similarity 0.85)
- Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective (NeurIPS 2024 · similarity 0.85)
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.
