Research map / Image generation / Image restoration
Early Exit Or Not: Resource-Efficient Blind Quality Enhancement for Compressed Images
Qunliang Xing, Mai Xu, Tianyi Li, Zhenyu Guan
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 RBQE instead of second-best CBDNet | blind enhancement · HEVC-MSP test average | Delta PSNR (dB): 0.409 -> 0.510 | Improved |
| use RBQE instead of DnCNN | blind enhancement · HEVC-MSP test average | Delta PSNR (dB): 0.317 -> 0.510 | Improved |
| use RBQE instead of QE-CNN | blind enhancement · HEVC-MSP test average | Delta PSNR (dB): 0.223 -> 0.510 | Improved |
| use RBQE instead of CBDNet | blind enhancement · JPEG test average | Delta PSNR (dB): 1.500 -> 1.678 | Improved |
| early-exit structure: exit at first decoding path instead of last for easiest samples (QP 22) (ablation) | RBQE · HEVC-MSP QP 22 | Delta PSNR (dB): 0.601 | No clear effect |
| exit at last decoding path instead of first for hardest samples (QP 42) (ablation) | RBQE · HEVC-MSP QP 42 | Delta PSNR (dB): +0.192 dB | Improved |
| force last exit rather than first exit for QP 22 samples (ablation) | RBQE · HEVC-MSP QP 22 | FLOPs (GMacs): 270% FLOPs | Hurt |
| increase quality threshold T_Q (ablation) | RBQE · HEVC-MSP | Delta PSNR (dB): saturates above 0.89 | Improved |
Closest papers by meaning
- Enhancing Quality of Compressed Images by Mitigating Enhancement Bias Towards Compression Domain (CVPR 2024 · similarity 0.86)
- DAQE: Enhancing the Quality of Compressed Images by Exploiting the Inherent Characteristic of Defocus (TPAMI 2023 · similarity 0.86)
- Breaking the Multi-Enhancement Bottleneck: Domain-Consistent Quality Enhancement for Compressed Images (TPAMI 2025 · similarity 0.86)
- Learning End-to-End Lossy Image Compression: A Benchmark (TPAMI 2021 · similarity 0.86)
- Make Lossy Compression Meaningful for Low-Light Images (AAAI 2024 · similarity 0.84)
- Unsupervised Blind Image Deblurring Based on Self-Enhancement (CVPR 2024 · similarity 0.84)
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.
