Research map / Image generation / Image restoration
Frequency-Aware Perceptual Optimization for Low-Complexity Implicit Image Compression
Haotian Wu, Gen Li, Di You, Pier Luigi Dragotti, Deniz Gunduz
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 |
|---|---|---|---|
| tune WA-WD subband weights (ablation) | Re2IC · Kodak medium bitrate | FID / DISTS: FID 40.01 -> 37.05; DISTS 0.0616 -> 0.0572 | Improved |
| tune WA-WD subband weights (ablation) | Re2IC · Kodak medium bitrate | PSNR (dB): 26.13 -> 25.65 | Hurt |
| region-wise coding with local-global modulation vs single-network overfitting (ablation) | Re2IC/WDs vs C3/WDs · Kodak | perceptual quality | Improved |
| add wavelet-Wasserstein control on top of region-wise WD (ablation) | Re2IC · Kodak | perceptual quality | Improved |
| optimize MSE alone instead of perceptual distortion (ablation) | Re2IC/mse · Kodak | perceptual quality | Hurt |
| partition image into salient regions with local networks (ablation) | Re2IC · encoding convergence (RD) | encoding steps to reach C3 BD-rate: 100k -> 25k steps | Improved |
| replace C3/WDs with Re2IC | Re2IC · CLIC2020 low bitrate | WD: 1.3082 -> 1.1131 | Improved |
| replace C3/WDs with Re2IC | Re2IC · CLIC2020 | PSNR: up to 0.82 dB gain | Improved |
| use Re2IC instead of HiFiC at medium bitrate | Re2IC · Kodak medium bitrate | PSNR (dB): 27.56 -> 26.13 | Hurt |
| use Re2IC instead of HiFiC | Re2IC · decoding complexity | decoding cost | Improved |
| use wavelet-decomposed WD (WA-WD) instead of WD as metric | WA-WD · IQA benchmarks | correlation with human ratings | Improved |
Closest papers by meaning
- Decouple Distortion from Perception: Region Adaptive Diffusion for Extreme-low Bitrate Perception Image Compression (CVPR 2025 · similarity 0.90)
- Towards image compression with perfect realism at ultra-low bitrates (ICLR 2024 · similarity 0.90)
- Controllable Distortion-Perception Tradeoff Through Latent Diffusion for Neural Image Compression (AAAI 2025 · similarity 0.89)
- Good, Cheap, and Fast: Overfitted Image Compression with Wasserstein Distortion (CVPR 2025 · similarity 0.88)
- Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion (ICML 2025 · similarity 0.87)
- Laplacian-guided Entropy Model in Neural Codec with Blur-dissipated Synthesis (CVPR 2024 · similarity 0.86)
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.
