13 datasets found

A large-scale dataset for human-aware motion deblurring, consisting of 8,422 pairs of blurry and sharp images with 65,784 densely annotated FG human bounding boxes.

The commonly used benchmark for image deblurring is Go-Pro (Nah, Kim, and Lee 2017), where the input blurry images are synthetic.

The dataset used in this paper is a collection of images from the STL-10 dataset, preprocessed and used for training and evaluation of the proposed diffusion spectral entropy...

Real-world camera motion blur and defocus blur datasets for training a sharp NeRF under blurry input.

Synthetic camera motion blur and defocus blur datasets for training a sharp NeRF under blurry input.

The Lai dataset is a benchmark for single image blind deblurring, containing 11 blurry images with no clean/sharp counterparts.

The Kohler dataset is a benchmark for blind image deblurring, containing 4 images with 12 different kernels.

The DVD dataset is a benchmark for image motion deblurring, containing 6708 synthetic blurry and sharp image pairs.

The GoPro dataset is used for training, which contains 2103 pairs of blurred clear control images. The validation set used for training is 460 pairs of randomly divided images....

The Helsinki Deblur Challenge 2021 dataset consists of 20 sub-problems of increasing defocus blur, with 203 pairs of sharp and blurred images for each level.

The dataset used in the paper is the synthetic defocus dataset, which consists of synthetic blurry training images and ground truth images.

The dataset used in the paper is the real defocus dataset, which consists of real-world blurry training images and ground truth images.

The dataset used in the paper is the Deblur-NeRF dataset, which consists of blurry training images and ground truth images.