Variational Denoising Network: Toward Blind Noise Modeling and Removal

Authors: Zongsheng Yue, Hongwei Yong, Qian Zhao, Deyu Meng, Lei Zhang

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Comprehensive experiments are performed to substantiate the superiority of our method in blind image denoising.
Researcher Affiliation Collaboration 1 School of Mathematics and Statistics, Xi an Jiaotong University, Shaanxi, China 2Department of Computing, Hong Kong Polytechnic University, Kowloon, Hong Kong 3DAMO Academy, Alibaba Group, Shenzhen, China 4Faculty of Information Technology, The Macau University of Science and Technology, Macau, China
Pseudocode No The paper does not include any pseudocode or clearly labeled algorithm blocks.
Open Source Code Yes The training and testing codes of our VDN is available at https://github.com/zsyOAOA/VDNet.
Open Datasets Yes We collected a set of source images to train the network, including 432 images from BSD [5], 400 images from the validation set of Image Net [12] and 4744 images from the Waterloo Exploration Database [26].
Dataset Splits Yes SIDD [1] is another real-world denoising benchmark, containing 30, 000 real noisy images... About 80% ( 24, 000 pairs) of this dataset are provided for training purpose, and the rest as held for benchmark. And 320 image pairs selected from them are packaged together as a medium version of SIDD... We employed this medium vesion dataset to train a real-world image denoiser, and test the performance on the two benchmarks. ... Table 5: Performance of VDN under different ε2 0 values on SIDD validation dataset (p = 7).
Hardware Specification No The paper does not provide specific hardware details such as GPU models, CPU types, or memory used for running the experiments.
Software Dependencies No The paper mentions "The Adam algorithm [21] is adopted to optimize the network parameters" but does not specify version numbers for any software libraries, frameworks, or programming languages used (e.g., Python, PyTorch, TensorFlow, CUDA versions).
Experiment Setup Yes The initial learning rate is set as 2e-4 and linearly decayed in half every 10 epochs until to 1e-6. The window size p in Eq. (3) is set as 7. The hyper-parameter ε2 0 is set as 5e-5 and 1e-6 in the following synthetic and real-world image denoising experiments, respectively.