Global Information Compensation Network for Image Denoising
Authors: Shifei Ding, Qidong Wang, Lili Guo
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results show that our proposed GICN effectively compensates for global information, achieves a balance between denoising efficiency and effect, and surpasses mainstream methods in multiple benchmark tests. |
| Researcher Affiliation | Academia | Shifei Ding1,2 , Qidong Wang1,2 , Lili Guo1,2 1School of Computer Science and Technology, China University of Mining and Technology 2Mine Digitization Engineering Research Center of Ministry of Education, China University of Mining and Technology EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper describes the methods through equations and network architecture diagrams (Figures 1-5), but does not contain explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain any explicit statements about the release of source code or links to a code repository. |
| Open Datasets | Yes | In this paper, we used a composite dataset for Gaussian denoising, consisting of 400 images from [Martin et al., 2001a], 800 from [Timofte et al., 2017], and 3000 from [Ma et al., 2016]. For real-world denoising, we trained on a combination of SIDD [Abdelhamed et al., 2018] and RENOIR [Anaya and Barbu, 2018]... In the experiment, we employed four datasets for testing synthetic noise: BSD68 [Martin et al., 2001b], Kodak24 [Russakovsky et al., 2015], Mc Master [Zhang et al., 2011], and CBSD68 [Martin et al., 2001b]. For real image denoising, we conducted testing using three datasets: SIDD [Abdelhamed et al., 2018], Nam [Nam et al., 2016], and Poly U [Xu et al., 2018]. |
| Dataset Splits | No | In this paper, we used a composite dataset for Gaussian denoising, consisting of 400 images from [Martin et al., 2001a], 800 from [Timofte et al., 2017], and 3000 from [Ma et al., 2016]. Grayscale and color versions were used separately to train corresponding denoising models. All images were cropped into 96 96 patches with random rotations for data augmentation. For real-world denoising, we trained on a combination of SIDD [Abdelhamed et al., 2018] and RENOIR [Anaya and Barbu, 2018], using 256 256 image patches. For testing synthetic noise: BSD68 [Martin et al., 2001b], Kodak24 [Russakovsky et al., 2015], Mc Master [Zhang et al., 2011], and CBSD68 [Martin et al., 2001b]. For real image denoising, we conducted testing using three datasets: SIDD [Abdelhamed et al., 2018], Nam [Nam et al., 2016], and Poly U [Xu et al., 2018]. While training and test datasets are mentioned, specific split percentages or counts for validation/test sets from the training pool are not detailed. |
| Hardware Specification | Yes | We implemented GICN using Py Torch 1.10 on an NVIDIA RTX 3090 GPU. |
| Software Dependencies | Yes | We implemented GICN using Py Torch 1.10 on an NVIDIA RTX 3090 GPU. |
| Experiment Setup | Yes | The model was trained for 100 epochs with a batch size of 128 using the Adam optimizer. The initial learning rate was set to 1e-4 and decayed to 1e-5 after 30 epochs. |