39 datasets found

The patch ordering method for image denoising uses √n by √n overlapping image patches.

The dataset used in the paper is a synthetic image dataset, consisting of 100 images with varying levels of noise.

The dataset used in the paper is a color image denoising dataset, consisting of 100 natural images with varying levels of noise.

The dataset used in this paper for contrast estimation and image denoising in cryo-electron microscopy.

The MNIST dataset is a large dataset of handwritten digits, and the Caltech101 Silhouettes dataset is a dataset of binary silhouettes of objects.

A three-dimensional retinal image denoising task using GraphLab

The SIDD and DND datasets are used for testing the CU-Mamba model.

The Urban100 dataset is a benchmark for image denoising, containing 100 images with varying levels of noise.

The dataset contains two sets of image patches each from the Experimental Data Record (i.e., Raw images without any corrections) and Reduced Data Record (i.e., corrected, and...

High-quality denoising dataset for smartphone cameras

The JPEG compression artifact removal dataset contains images with strong JPEG compression artifacts.

The dataset used in the paper is a synthetic dataset for blind super-resolution, consisting of low-resolution images and their corresponding blur kernels.

The dataset used for Gaussian denoising, created by augmenting and frequency-manipulating images to tap into the epistemic uncertainty of a pretrained denoiser.

The dataset used in the paper is McMaster, a dataset for image denoising.

The dataset used in the paper for image denoising and image restoration tasks.

Two public low-dose CT denoising datasets

Anatomy-aware Supervised Contrastive Learning Framework for Low-dose CT Denoising

The House dataset contains house sale prices for the King County area. The prediction task is inferring whether a house costs at least the median house price.

We hold out 3% of the data for validation and 5% for testing. For additional test data, we render a few PBRT-v3 test scenes that portray practical real-world scenarios1 though...

Rendering We generate approx. 5650 random scenes using a procedure similar to [21]. Each scene contains randomly arranged ShapeNet objects, textures, and lighting configurations...