Masked Pre-training Enables Universal Zero-shot Denoiser
Authors: Xiaoxiao Ma, Zhixiang Wei, Yi Jin, Pengyang Ling, Tianle Liu, Ben Wang, Junkang Dai, Huaian Chen
NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Comprehensive experiments across various noisy scenarios underscore the notable advances of MPI over previous approaches with a marked reduction in inference time. Code available at https://github.com/krennic999/MPI. and 3 Experiments We assess our method against typical methods including DIP [13], Noise2Void (N2V) [16], Noise2Self (N2S) [17], Zero-Shot Noise2Noise (ZS-N2N) [22], and Faster DIP [19]. |
| Researcher Affiliation | Academia | Xiaoxiao Ma1 Zhixiang Wei1 Yi Jin1 Pengyang Ling1,2 Tianle Liu1 Ben Wang1 Junkang Dai1 Huaian Chen1 1 University of Science and Technology of China 2 Shanghai AI Laboratory {xiao_xiao,zhixiangwei,lpyang27,tleliu,wblzgrsn,junkangdai,anchen}@mail.ustc.edu.cn {jinyi08}@ustc.edu.cn |
| Pseudocode | Yes | Algorithm 1: Iterative filling. Pipeline designed to leverage pre-trained representation θ for zero-shot denoising. Input: Noisy image x, pre-trained parameter θ, network D( ), exponential weight β, masking ratio p. Output: denoised ensemble y from predictions of iteration {yt}. |
| Open Source Code | Yes | Code available at https://github.com/krennic999/MPI. |
| Open Datasets | Yes | Pre-training. Pre-training is performed on two Nvidia RTX 3090 GPUs using Adam optimizer with β1=0.9 and β2=0.9. Initial learning rate is 2e 3 and decays to 1e 5 with cosine annealing strategy over 80K iterations with a batch size of 64. We initiate pre-training on randomly cropped 256 256 patches from subset of Image Net [28] with around 48,000 images. We investigate Gaussian Noise with σ [10,25,50] and Poisson noise with λ [10, 25, 50] separately on three datasets: CSet [3], Mc Master [33] and CBSD [34], with 9, 18 and 68 high-quality images, respectively. |
| Dataset Splits | Yes | We test on SIDD [1] and Poly U [45] datasets, including 1280 patches from the SIDD validation and 1280 from SIDD benchmark, and all 100 official patches from Poly U to show our paradigm on real images. |
| Hardware Specification | Yes | Pre-training. Pre-training is performed on two Nvidia RTX 3090 GPUs using Adam optimizer with β1=0.9 and β2=0.9. |
| Software Dependencies | No | The paper mentions ‘Adam optimizer’ and ‘U-shaped hourglass architecture as in DIP [13]’ but does not provide specific version numbers for any software, libraries, or frameworks used for implementation or experimentation. |
| Experiment Setup | Yes | Pre-training. Pre-training is performed on two Nvidia RTX 3090 GPUs using Adam optimizer with β1=0.9 and β2=0.9. Initial learning rate is 2e 3 and decays to 1e 5 with cosine annealing strategy over 80K iterations with a batch size of 64. ... Zero-shot inference. We set learning rate during inference to 2e 3, and same masking ratio p as pre-training (0.3 for synthetic, 0.8 0.95 for real noise) is set. EMA weight β=0.99 for 1000 iterations (specially, 800 iterations for SIDD). Additionally, with β=0.9, we achieve performance surpassing most zero-shot methods within 200 iterations, denoted as faster . |