Research map / Image generation / Super-resolution
Contourlet Refinement Gate Framework for Thermal Spectrum Distribution Regularized Infrared Image Super-Resolution
Yang Zou, Zhixin Chen, Z. Y. Zhang, Xingyuan Li, Long Ma, Jinyuan Liu, Peng Wang, Yanning zhang
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 Directional Filter Bank and Laplacian Pyramid to the Contourlet block (ablation) | Contourlet Refinement Gate · M3FD x4 SR | PSNR (dB): 40.031 -> 40.787 | Improved |
| add Laplacian Pyramid to DFB-only Contourlet block (ablation) | Contourlet block · M3FD x4 SR | PSNR (dB): 40.469 -> 40.787 | Improved |
| add Directional Filter Bank to LP-only Contourlet block (ablation) | Contourlet block · M3FD x4 SR | PSNR (dB): 40.521 -> 40.787 | Improved |
| increase Contourlet decomposition levels from 1 to 4 (ablation) | Contourlet block · M3FD x4 SR | PSNR (dB): 40.742 -> 40.787 | Improved |
| increase Contourlet decomposition levels from 1 to 4 (ablation) | Contourlet block · M3FD x4 SR | SSIM: 0.9652 -> 0.9774 | Mixed |
| add positive prompt to the baseline (ablation) | prompt-guided SR · M3FD x4 SR | PSNR (dB): 40.431 -> 40.742 | Improved |
| add negative prompt on top of positive prompt (ablation) | prompt-guided SR · M3FD x4 SR | PSNR (dB): 40.742 -> 40.787 | Improved |
| add negative prompt on top of positive prompt (ablation) | prompt-guided SR · M3FD x4 SR | SSIM: 0.9653 -> 0.9774 | Improved |
| use two-stage learned prompts instead of fixed prompts (ablation) | prompt-guided SR · M3FD x4 SR | PSNR (dB): 40.662 -> 40.787 | Improved |
| learn both prompts instead of only the negative prompt (ablation) | prompt-guided SR · M3FD x4 SR | PSNR (dB): 40.725 -> 40.787 | Improved |
| use the proposed Contourlet Refinement Gate SR instead of HAT-L at x4 on Set15 | infrared SR · M3FD Set15 (x4 SR) | PSNR (dB): 40.741 -> 42.741 | Improved |
| feed proposed SR outputs to YOLOv5 instead of the best RGB SR (DAT) at x2 | downstream detection · M3FD detection (x2 SR inputs) | mAP@.5:.95: 0.388 -> 0.408 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ResShift — ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting (NeurIPS 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Contourlet Residual for Prompt Learning Enhanced Infrared Image Super-Resolution (ECCV 2024 · similarity 0.93)
- Toward Real-world Infrared Image Super-Resolution: A Unified Autoregressive Framework and Benchmark Dataset (CVPR 2026 · similarity 0.90)
- DifIISR: A Diffusion Model with Gradient Guidance for Infrared Image Super-Resolution (CVPR 2025 · similarity 0.88)
- Enhancing Infrared Vision: Progressive Prompt Fusion Network and Benchmark (NeurIPS 2025 · similarity 0.86)
- Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts (CVPR 2024 · similarity 0.85)
- Block-Based Multi-Scale Image Rescaling (AAAI 2025 · 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.
