Research map / Architectures and efficiency / Neural architecture search
Memory-Efficient Hierarchical Neural Architecture Search for Image Restoration
Haokui Zhang, Ying Li, Hao Chen, Chengrong Gong, Zongwen Bai, Chunhua Shen
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 layer-wise architecture sharing (ablation) | HiNAS · BSD500 denoising (sigma=30) | PSNR: 29.15 -> 29.17 | Improved |
| add residual learning (ablation) | HiNAS · BSD500 denoising (sigma=30) | PSNR: 29.15 -> 29.20 | Improved |
| add both LWAS and residual learning (ablation) | HiNAS · BSD500 denoising (sigma=30) | PSNR: 29.15 -> 29.23 | Improved |
| add residual learning (ablation) | HiNAS · BSD100 x4 super-resolution | PSNR: 26.92 -> 27.32 | Improved |
| add layer-wise architecture sharing (ablation) | HiNAS · BSD100 x4 super-resolution | PSNR: 26.92 -> 27.03 | Improved |
| add both strategies (ablation) | HiNAS · BSD100 x4 super-resolution | PSNR: 26.92 -> 27.35 | Improved |
| search the outer layer width (w40 instead of fixed width) (ablation) | HiNAS · BSD500 denoising | PSNR: 29.25 -> 29.14 | Hurt |
| add l_ssim loss to MSE (ablation) | HiNAS · BSD500 denoising (sigma=30) | SSIM: 0.8254 -> 0.8403 | Improved |
| add l_ssim loss to MSE (ablation) | HiNAS · BSD100 x4 super-resolution | PSNR: 27.38 -> 27.35 | Hurt |
| add l_ssim loss to MSE (ablation) | HiNAS · BSD100 x4 super-resolution | SSIM: 0.7367 -> 0.7467 | Improved |
| replace searched operations with other convolutions (R1) (ablation) | HiNAS · BSD500 denoising | PSNR: 29.25 -> 29.21 | Hurt |
| change cell topology (R2) (ablation) | HiNAS · BSD500 denoising | SSIM: 0.8420 -> 0.8304 | Hurt |
Closest papers by meaning
- Memory-Efficient Hierarchical Neural Architecture Search for Image Denoising (CVPR 2020 · similarity 0.97)
- CLEARER: Multi-Scale Neural Architecture Search for Image Restoration (NeurIPS 2020 · similarity 0.92)
- ISNAS-DIP: Image-Specific Neural Architecture Search for Deep Image Prior (CVPR 2022 · similarity 0.90)
- NAS-DIP: Learning Deep Image Prior with Neural Architecture Search (ECCV 2020 · similarity 0.90)
- Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective (ICLR 2021 · similarity 0.89)
- Efficient Few-Shot Neural Architecture Search by Counting the Number of Nonlinear Functions (AAAI 2025 · similarity 0.88)
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.
