Noisier2Noise: Learning to Denoise From Unpaired Noisy Data

Moran, Nick; Schmidt, Dan; Zhong, Yu; Chiu, Patrick

doi:10.1109/cvpr42600.2020.01208

Cited by 164 publications

(124 citation statements)

References 10 publications

Supporting

Mentioning

124

Contrasting

Order By: Relevance

“…Within the Refine step of IsoNet, we also implemented a denoising module based on the noisier-input strategy 20,21 . When this optional denoising module is enabled in the Refine step, in each iteration, 3D noise volumes are reconstructed by the back-projection algorithm from a series of 2D images containing only Gaussian noise.…”

Section: Refinementioning

confidence: 99%

See 1 more Smart Citation

Isotropic Reconstruction of Electron Tomograms with Deep Learning

Liu

Zhang

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

Cryogenic electron tomography (cryoET) allows visualization of cellular structures in situ. However, anisotropic resolution arising from the intrinsic missing-wedge problem has presented major challenges in visualization and interpretation of tomograms. Here, we have developed IsoNet, a deep learning-based software package that iteratively reconstructs the missing-wedge information and increases signal-to-noise ratio, using the knowledge learned from raw tomograms. Without the need for sub-tomogram averaging, Isonet generates tomograms with significantly reduced resolution anisotropy. Applications of IsoNet to three representative types of cryoET data demonstrate greatly improved structural interpretability: resolving lattice defects in immature HIV particles, establishing architecture of the paraflagellar rod in Eukaryotic flagella, and identifying heptagon-containing clathrin cages inside a neuronal synapse of cultured cells. Therefore, by overcoming two fundamental limitations of cryoET, IsoNet enables functional interpretation of cellular tomograms without sub-tomogram averaging. Its application to high-resolution cellular tomograms should also help identify differently oriented complexes of the same kind for near-atomic resolution sub-tomogram averaging.

show abstract

Section: Refinementioning

confidence: 99%

“…This optional step allows performing missing-wedge correction and denoising simultaneously using IsoNet. IsoNet uses a noisier-input strategy 20,21 that learns to map "input" with additional noise to the "target".…”

Section: Refine Step 2: Adding Noisementioning

confidence: 99%

Isotropic Reconstruction of Electron Tomograms with Deep Learning

Liu

Zhang

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…For training denoisers, the lack of clean image data to be used as groundtruth may be a challenge for certain applications like medical imaging and remote sensing [22,35,47,50]. In this scenario, different training strategies were proposed.…”

Section: Full Research Papermentioning

confidence: 99%

“…In this scenario, different training strategies were proposed. The Noise2Noise [24] and Noisier2Noise [35] approaches consist in the idea of training with pairs of noisy images, where the clean image can be predicted by learning common patterns in both images which are supposed to be present in the clean image. There is also the Noise2Void [21] where the learning process is done with only corrupted or noisy images, and the noisy pattern is learned considering the given dataset.…”

Section: Full Research Papermentioning

confidence: 99%

A Denoising Convolutional Neural Network for Self-Supervised Rank Effectiveness Estimation on Image Retrieval

Valem

Pedronette

2021

Proceedings of the 2021 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

Image and multimedia retrieval has established as a prominent task in an increasingly digital and visual world. Mainly supported by decades of development on hand-crafted features and the success of deep learning techniques, various different feature extraction and retrieval approaches are currently available. However, the frequent requirements for large training sets still remain as a fundamental bottleneck, especially in real-world and large-scale scenarios. In the scarcity or absence of labeled data, choosing what retrieval approach to use became a central challenge. A promising strategy consists in to estimate the effectiveness of ranked lists without requiring any groundtruth data. Most of the existing measures exploit statistical analysis of the ranked lists and measure the reciprocity among lists of images in the top positions. This work innovates by proposing a new and self-supervised method for this task, the Deep Rank Noise Estimator (DRNE). An algorithm is presented for generating synthetic ranked list data, which is modeled as images and provided for training a Convolutional Neural Network that we propose for effectiveness estimation. The proposed model is a variant of the DnCNN (Denoiser CNN), which intends to interpret the incorrectness of a ranked list as noise, which is learned by the network. Our approach was evaluated on 5 public image datasets and different tasks, including general image retrieval and person re-ID. We also exploited and evaluated the complementary between the proposed approach and related rank-based approaches through fusion strategies. The experimental results showed that the proposed method is capable of achieving up to 0.88 of Pearson correlation with MAP measure in general retrieval scenarios and 0.74 in person re-ID scenarios.

show abstract

“…While most of the DCNN models are trained using pairs of noisy and clean images, some of the recent methods, such as Noise2Void (Krull et al 2019a), Noise2Self (Batson and Royer 2019) can be unsupervised, but at a price of degraded performance ). Noiser2Noise (Moran et al 2020), probabilistic Noise2Void and parametric probabilistic Noise2Void (Prakash et al 2020) (PPN2V) improve the performance by introducing estimated noise models.…”

Section: Introductionmentioning

confidence: 99%

Noise2Atom: unsupervised denoising for scanning transmission electron microscopy images

et al. 2020

View full text Add to dashboard Cite

We propose an effective deep learning model to denoise scanning transmission electron microscopy (STEM) image series, named Noise2Atom, to map images from a source domain $\mathcal {S}$ S to a target domain $\mathcal {C}$ C , where $\mathcal {S}$ S is for our noisy experimental dataset, and $\mathcal {C}$ C is for the desired clear atomic images. Noise2Atom uses two external networks to apply additional constraints from the domain knowledge. This model requires no signal prior, no noise model estimation, and no paired training images. The only assumption is that the inputs are acquired with identical experimental configurations. To evaluate the restoration performance of our model, as it is impossible to obtain ground truth for our experimental dataset, we propose consecutive structural similarity (CSS) for image quality assessment, based on the fact that the structures remain much the same as the previous frame(s) within small scan intervals. We demonstrate the superiority of our model by providing evaluation in terms of CSS and visual quality on different experimental datasets.

show abstract

Noisier2Noise: Learning to Denoise From Unpaired Noisy Data

Cited by 164 publications

References 10 publications

Isotropic Reconstruction of Electron Tomograms with Deep Learning

Isotropic Reconstruction of Electron Tomograms with Deep Learning

A Denoising Convolutional Neural Network for Self-Supervised Rank Effectiveness Estimation on Image Retrieval

Noise2Atom: unsupervised denoising for scanning transmission electron microscopy images

Contact Info

Product

Resources

About