Spectrum-to-Kernel Translation for Accurate Blind Image Super-Resolution
Authors: Guangpin Tao, Xiaozhong Ji, Wenzhuo Wang, Shuo Chen, Chuming Lin, Yun Cao, Tong Lu, Donghao Luo, Ying Tai
NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on both synthetic and real-world images demonstrate that our proposed method sufficiently reduces blur kernel estimation error, thus enables the off-the-shelf non-blind SR methods to work under blind setting effectively, and achieves superior performance over state-of-the-art blind SR methods, averagely by 1.39d B, 0.48d B on commom blind SR setting (with Gaussian kernels) for scales 2 and 4 , respectively. |
| Researcher Affiliation | Collaboration | 1National Key Lab for Novel Software Technology, Nanjing University 2Tencent Youtu Lab 3RIKEN Center for Advanced Intelligence Project |
| Pseudocode | No | The paper describes the S2K network structure and its components but does not include structured pseudocode or an algorithm block. |
| Open Source Code | No | The paper does not explicitly state that source code for the described methodology is released or provide a link to it. |
| Open Datasets | Yes | We use DIV2K [28] training set to train S2K and the non-blind SR model. |
| Dataset Splits | Yes | We use DIV2K [28] training set to train S2K and the non-blind SR model. Then, we use random blurred DIV2K validation set, 100 images in Flicker2K [1] as test datasets in the synthetic experiment, and conduct 2 , 3 , 4 SR comparison and kernel estimation accuracy comparison. |
| Hardware Specification | No | No specific hardware details (e.g., CPU, GPU models, or detailed computer specifications) used for running experiments were mentioned in the paper. |
| Software Dependencies | No | The paper mentions using Adam for optimization but does not provide specific version numbers for software dependencies like programming languages, libraries, or frameworks. |
| Experiment Setup | Yes | We set the weights of loss item in generator as λ1 = 100, λ2 = 1, λ3 = 1 respectively. For optimization, we use Adam with β1 = 0.5, β2 = 0.999, and learning rate is 0.001. |