An in-the-wild stereo image dataset, comprising 49,368 image pairs contributed by users of the Holopix mobile social platform. Source: Holopix50k: A Large-Scale In-the-wild Stereo Image Dataset

A dataset consisting of stereo thermal, stereo color, and cross-modality image pairs with high accuracy ground truth (< 2mm) generated from a LiDAR. The authors scanned 100 cluttered indoor and 80 outdoor scenes featuring challenging environments and conditions. CATS contains approximately 1400 images of pedestrians, vehicles, electronics, and other thermally interesting objects in different environmental conditions, including nighttime, daytime, and foggy scenes. Source: CATS: A Color and Thermal Stereo Benchmark

RGB-D-D is a large-scale dataset for depth map super-resolution (SR). It consists of real-world paired low-resolution (LR) and high-resolution (HR) depth maps. The paired LR and HR depth maps are captured from mobile phone and Lucid Helios respectively ranging from indoor scenes to challenging outdoor scenes.

QMUL-SurvFace is a surveillance face recognition benchmark that contains 463,507 face images of 15,573 distinct identities captured in real-world uncooperative surveillance scenes over wide space and time. Source: Surveillance Face Recognition Challenge

Middlebury 2005 is a stereo dataset of indoor scenes. Source: https://vision.middlebury.edu/stereo/data/scenes2005/ Image Source: https://vision.middlebury.edu/stereo/data/scenes2005/

Nearly 10,000 km² of free high-resolution and paired multi-temporal low-resolution satellite imagery of unique locations which ensure stratified representation of all types of land-use across the world: from agriculture to ice caps, from forests to multiple urbanization densities. ​ Those locations are also enriched with typically under-represented locations in ML datasets: sites of humanitarian interest, illegal mining sites, and settlements of persons at risk. ​ Each high-resolution image (Airbus SPOT at up to 1.5 m/pixel) comes with multiple temporally-matched low-resolution images from the freely accessible lower-resolution Sentinel-2 satellites (up to 10 m/pixel, 12 spectral bands). ​ The dataset is accompanied with a paper, datasheet for datasets and an open-source Python package to: rebuild or extend the WorldStrat dataset, train and infer baseline algorithms, and learn with abundant tutorials, all compatible with the popular EO-learn toolbox. The hope is to foster broad-spectrum applications of ML to satellite imagery, and possibly develop the same power of analysis allowed by costly private high-resolution imagery from free public low-resolution Sentinel2 imagery. We illustrate this specific point by training and releasing several highly compute-efficient baselines on the task of Multi-Frame Super-Resolution.

OST300 is an outdoor scene dataset with 300 test images of outdoor scenes, and a training set of 7 categories of images with rich textures. Source: Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform Image Source: http://mmlab.ie.cuhk.edu.hk/projects/SFTGAN/

The PROBA-V Super-Resolution dataset is the official dataset of ESA's Kelvins competition for "PROBA-V Super Resolution". It contains satellite data from 74 hand-selected regions around the globe at different points in time. The data is composed of radiometrically and geometrically corrected Top-Of-Atmosphere (TOA) reflectances for the RED and NIR spectral bands at 300m and 100m resolution in Plate Carrée projection. The 300m resolution data is delivered as 128x128 grey-scale pixel images, the 100m resolution data as 384x384 grey-scale pixel images. Additionally, a quality map is provided for each pixel, indicating whether the pixels are concealed (i.e. by clouads, ice, water, missing information, etc.). The goal of the challenge can be described as Multi-Image Super-resolution: Construct a single high-resolution image out of a series of more frequent low resolution images. Detailed information about the related competition can be found at https://kelvins.esa.int/proba-v-super-resolution.

The Stanford Light Field Archive is a collection of several light fields for research in computer graphics and vision. Source: http://lightfield.stanford.edu/ Image Source: http://lightfield.stanford.edu/

We present a novel approach to reference-based super-resolution (RefSR) with the focus on real-world dual-camera super-resolution (DCSR). This dataset currently consists of 143 pairs of telephoto and wide-angle images in 4K resolution captured by smartphone dual-cameras. See our paper for more details: Dual-Camera Super-Resolution with Aligned Attention Modules.

AeroRIT is a hyperspectral dataset to facilitate aerial hyperspectral scene understanding. Source: AeroRIT: A New Scene for Hyperspectral Image Analysis Image Source: https://github.com/aneesh3108/AeroRIT

Our RealMCVSR dataset provides real-world HD video triplets concurrently recorded by Apple iPhone 12 Pro Max equipped with triple cameras having fixed focal lengths: ultra-wide (30mm), wide-angle (59mm), and telephoto (147mm). To concurrently record video triplets, we built an iOS app that provides full control over exposure parameters (i.e., shutter speed and ISO) of the cameras. For recording each scene, we set the cameras in the auto-exposure mode, where the shutter speeds of the three cameras are synced to avoid varying motion blur across a video triplet. ISOs are adjusted accordingly for each camera to pick up the same exposure. Each video is saved in the MOV format using HEVC/H.265 encoding with the HD resolution (1080 x 1920). The dataset contains triplets of 161 video clips with 23,107 frames in total. The video triplets are split into training, validation, and testing sets, each of which has 137, 8, and 16 triplets with 19,426, 1,141, and 2,540 frames, respectively.

QST contains 1,167 video clips that are cut out from 216 time-lapse 4K videos collected from YouTube, which can be used for a variety of tasks, such as (high-resolution) video generation, (high-resolution) video prediction, (high-resolution) image generation, texture generation, image inpainting, image/video super-resolution, image/video colorization, image/video animating, etc. Each short clip contains multiple frames (from a minimum of 58 frames to a maximum of 1,200 frames, a total of 285,446 frames), and the resolution of each frame is more than 1,024 x 1,024. Specifically, QST consists of a training set (containing 1000 clips, totally 244,930 frames), a validation set (containing 100 clips, totally 23,200 frames), and a testing set (containing 67 clips, totally 17,316 frames). Click here (Key: qst1) to download the QST dataset.

This dataset presents a vision and perception research dataset collected in Rome, featuring RGB data, 3D point clouds, IMU, and GPS data. We introduce a new benchmark targeting visual odometry and SLAM, to advance the research in autonomous robotics and computer vision. This work complements existing datasets by simultaneously addressing several issues, such as environment diversity, motion patterns, and sensor frequency. It uses up-to-date devices and presents effective procedures to accurately calibrate the intrinsic and extrinsic of the sensors while addressing temporal synchronization. During recording, we cover multi-floor buildings, gardens, urban and highway scenarios. Combining handheld and car-based data collections, our setup can simulate any robot (quadrupeds, quadrotors, autonomous vehicles). The dataset includes an accurate 6-dof ground truth based on a novel methodology that refines the RTK-GPS estimate with LiDAR point clouds through Bundle Adjustment. All sequences divided in training and testing are accessible through our website.

Alsat-2B is a remote sensing dataset of low and high spatial resolution images (10m and 2.5m respectively) for the single-image super-resolution task. The high-resolution images are obtained through pan-sharpening. The dataset has been created from 13 images captured by the Alsat-2B Earth observation satellite, where the image cover 13 different cities.

The airborne hyperspectral dataset was taken by Headwall Hyperspec-VNIR-C imaging sensor over agricultural and urban areas in Chikusei, Ibaraki, Japan, on July 29, 2014 between the times 9:56 to 10:53 UTC+9. The central point of the scene is located at coordinates: 36.294946N, 140.008380E. The hyperspectral dataset has 128 bands in the spectral range from 363 nm to 1018 nm. The scene consists of 2517x2335 pixels and the ground sampling distance was 2.5 m. Ground truth of 19 classes was collected via a field survey and visual inspection using high-resolution color images obtained by Canon EOS 5D Mark II together with the hyperspectral data. The hyperspectral data and ground truth were made available to the scientific community in the ENVI and MATLAB formats at http://park.itc.u-tokyo.ac.jp/sal/hyperdata. More details of the experiment are presented in the technical report given below. In order to use the datasets, please fulfill the following three requirements: 1) Giving an acknowledgement as follows: The authors gratefully acknowledge Space Application Laboratory, Department of Advanced Interdisciplinary Studies, the University of Tokyo for providing the hyperspectral data. 2) Using the following license for hyperspectral data: http://creativecommons.org/licenses/by/3.0/ 3) This dataset was made public by Dr. Naoto Yokoya and Prof. Akira Iwasaki from the University of Tokyo. Please cite: In WORD: N. Yokoya and A. Iwasaki, "Airborne hyperspectral data over Chikusei," Space Appl. Lab., Univ. Tokyo, Japan, Tech. Rep. SAL-2016-05-27, May 2016. In LaTex: @techreport{NYokoya2016, author = {N. Yokoya and A. Iwasaki}, title = {Airborne hyperspectral data over Chikusei}, institution = {Space Application Laboratory, University of Tokyo}, number = {SAL-2016-05-27}, address = {Japan}, month = {May}, year = 2016, }

A large-scale multi-scene dataset for stereo deblurring, containing 20,637 blurry-sharp stereo image pairs from 135 diverse sequences and their corresponding bidirectional disparities. Source: DAVANet: Stereo Deblurring with View Aggregation

EDFace-Celeb-1M is a public Ethnically Diverse Face dataset which is used to benchmark the task of face hallucination. The dataset includes 1.7 million photos that cover different countries, with balanced race composition.

Real-CE is a real-world Chinese-English benchmark dataset for the task of STISR with the emphasis on restoring structurally complex Chinese characters. The benchmark provides 1,935/783 real LR-HR text image pairs (contains 33,789 text lines in total) for training/testing in 2× and 4× zooming modes, complemented by detailed annotations, including detection boxes and text transcripts.

SEPE 8K dataset is made of 40 different 8K (8192 x 4320) video sequences and 40 variant 8K (8192 x 5464) images. The video sequences were captured at a framerate of 29.97 frames per second (FPS) and had been encoded into videos using AVC/H.264, HEVC/H.265, and AV1 codecs at resolutions from 8K to 480p. The images, video sequences, encoded videos, and various other statistics related to the media that make the dataset are stored online, published, and maintained on the repo on GitHub for non-commercial use. this proposed dataset is - as far as we know - the first to publish true 8K natural sequences; thus, it is important for the next level of applications dealing with multimedia such as video quality assessment, super-resolution, video coding, video compression, and many more. Paper: https://doi.org/10.1145/3587819.3592560

StereoMSI comprises of 350 registered colour-spectral image pairs. The dataset has been used for the two tracks of the PIRM2018 challenge. Source: PIRM2018 Challenge on Spectral Image Super-Resolution: Dataset and Study Image Source: Shoeiby et al

Digitally Generated Numerals (DIGITal) ## Description The Digitally Generated Numerals (DIGITal) dataset consists of 100,000 image pairs representing digits from 0 to 9. These image pairs include both low and high-quality versions, with a resolution of 128x128 pixels. ## Languages The dataset is generated from fonts in the following languages: - Arabic - Chinese (Simplified) - Farsi and Urdu - Gurmukhi - Gujarati - Tibetan - ARDIS (Sweden) - ISI Bangla - Bangla Lekha - Kannada - English ## Image Resolution - resolution: 128x128 pixels ## Image Pairs - Each digit (0 to 9) has 1000 associated images. ## Total Images - Total images: 100,000 ## Samples | !Image 1 | !Image 2 | !Image 3 | !Image 4 | !Image 5 | |:-----------------------:|:-----------------------:| | Arabic | ARDIS | Chinese | Farsi & Urdu | Gujarati | | !Image 5 | !Image 6 | !Image 7 | !Image 8 | !Image 5 | |:-----------------------:|:-----------------------:| | ISI Bangla & BanglaLekha | Kannada | Tibetan | English | Gurmukhi This dataset is a valuable resource for tasks related to digit recognition, font analysis, and super-resolution. Researchers and practitioners can utilize it for various applications in computer vision and machine learning.

The INRIA Dense Light Field Dataset (DLFD) is a dataset for testing depth estimation methods in a light field. DLFD contains 39 scenes with disparity range [-4,4] pixels. The light fields are of spatial resolution 512 x 512 and angular resolution 9 x 9. Source: http://clim.inria.fr/Datasets/InriaSynLF/index.html Image Source: http://clim.inria.fr/Datasets/InriaSynLF/index.html

The Sentinel-2 satellite carries 12 CMOS detectors for the VNIR bands, with adjacent detectors having overlapping fields of view that result in overlapping regions in level-1 B (L1B) images. This dataset includes 3740 pairs of overlapping image crops extracted from two L1B products. Each crop has a height of around 400 pixels and a variable width that depends on the overlap width between detectors for RGBN bands, typically around 120-200 pixels. In addition to detector parallax, there is also cross-band parallax for each detector, resulting in shifts between bands. Pre-registration is performed for both cross-band and cross-detector parallax, with a precision of up to a few pixels (typically less than 10 pixels).

With the global availability of Multispectral Instrument (MSI, on-board Sentinel-2 satellite) data and Operational Land Imager (OLI, on-board Landsat-8 satellite) data which have 10 m and 30 m ground sample distance (GSD), respectively, it is conceivable that we can construct a dataset of LR-HR image pairs with OLI data as the LR image and MSI data as the HR image. This dataset, termed OLI2MSI, is a collection of paired real-world multisensor LR-HR data by carefully selecting relatively cloud-free MSI and OLI image in the same location obtained within a suitable temporal window.

Advanced pixel shift technology is employed to perform a full color sampling of the image. Pixel shift technology takes four samples of the same image at nearly the same time, and physically controls the camera sensor to move one pixel horizontally or vertically at each sampling to capture all color information at each pixel. The pixel shift technology ensures that the sampled images follow the distribution of natural images sampled by the camera, and the full information of the color (R, Gr, Gb, B channel) is completely obtained without any need of interpolation. In this way, the collected RGB images are artifacts-free, which leads to better training results for demosaicing related tasks. PixelShift200 Dataset contains 210 high quality 4K images. - Training: 200 images - Testing: 10 images - Key Features: fully colored, demosiacing artifacts free - Camera: SONY α7R III Source: PixelShift200

SAT-MTB-VSR is a large-scale dataset for satellite video super-resolution made from original videos of Jilin-1, which is a subset of the satellite video multitasking dataset SAT-MTB. The dataset is cropped from 18 videos captured by the Jilin-1 video satellite, covering a wide range of terrains, such as cities, docks, airports, suburbs, forests, and deserts, with a resolution of about 1 m. And the videos contain dynamic scenes, such as moving cars, airplanes, trains, and ships, which test the ability of the VSR method to deal with moving targets of different sizes and speeds. At the same time, due to the motion of the satellite, the video contains changes in viewing angle and lighting. It contains 431 videos, each of which is 100 consecutive frames, of which 413 are used as the training set and 18 as the validation set, and all the 18 validation sets are from different original videos, while the size of the images is 640 × 640. These images are downsampled 4× by bicubic interpolation to get 160 × 160 low-resolution images, thus obtaining the LR-HR training pairs.

SEN2VENµS is an open dataset for the super-resolution of Sentinel-2 images by leveraging simultaneous acquisitions with the VENµS satellite. The dataset is composed of 10m and 20m cloud-free surface reflectance patches from Sentinel-2, with their reference spatially-registered surface reflectance patches at 5 meters resolution acquired on the same day by the VENµS satellite. This dataset covers 29 locations with a total of 132 955 patches of 256x256 pixels at 5 meters resolution, and can be used for the training of super-resolution algorithms to bring spatial resolution of 8 of the Sentinel-2 bands down to 5 meters.

To construct such a dataset, a straightforward approach was scraping images from the web. The main source for our dataset is ShipSpotting, which serves as a repository for user uploaded images, hosting a vast collection of ship images, amounting to approximately 3 million. Furthermore, for each image, valuable supplementary information is available, such as the type of the ship, and present and past names. Next, we made sure that as many images as possible were collected in our dataset, since in deep learning, the quantity of training data directly influences the quality of results. A larger volume of data enables models to generalize more effectively. Thus we scrape all the images and as a result, the dataset comprises a total of 1.517.702 samples. We exclude many classes of ships from our final analysis and concentrate on the more common and valuable for a real scenario use case. The total number of different classes is 20 and the ship categories included are Bulkers, Containerships, Cruise ships, Dredgers, Fire Fighting Vessels, Floating Sheerlegs, General Cargo, Inland, Livestock Carriers, Passenger Vessels, Patrol Forces, Reefers, Ro-ro, Supply ships, Tankers, Training ships, Tugs, Vehicle Carriers, Wood Chip Carriers. The total amount of samples after this class selection is 507.918.

The availability of well-curated datasets has driven the success of Machine Learning (ML) models. Despite greater access to earth observation data in agriculture, there is a scarcity of curated and labelled datasets, which limits the potential of its use in training ML models for remote sensing (RS) in agriculture. To this end, we introduce a first-of-its-kind dataset called SICKLE, which constitutes a time-series of multi-resolution imagery from 3 distinct satellites: Landsat-8, Sentinel-1 and Sentinel-2. Our dataset constitutes multi-spectral, thermal and microwave sensors during January 2018 - March 2021 period. We construct each temporal sequence by considering the cropping practices followed by farmers primarily engaged in paddy cultivation in the Cauvery Delta region of Tamil Nadu, India; and annotate the corresponding imagery with key cropping parameters at multiple resolutions (i.e. 3m, 10m and 30m). Our dataset comprises 2, 370 season-wise samples from 388 unique plots, having an average size of 0.38 acres, for classifying 21 crop types across 4 districts in the Delta, which amounts to approximately 209, 000 satellite images. Out of the 2, 370 samples, 351 paddy samples from 145 plots are annotated with multiple crop parameters; such as the variety of paddy, its growing season and productivity in terms of per-acre yields. Ours is also one among the first studies that consider the growing season activities pertinent to crop phenology (spans sowing, transplanting and harvesting dates) as parameters of interest. We benchmark SICKLE on three tasks: crop type, crop phenology (sowing, transplanting, harvesting), and yield prediction.

Confocal fluorescence microscopy is one of the most accessible and widely used imaging techniques for the study of biological processes at the cellular and subcellular levels. Scanning confocal microscopy allows the capture of high-quality images from thick three-dimensional (3D) samples, yet suffers from well-known limitations such as photobleaching and phototoxicity of specimens caused by intense light exposure, which limits its use in some applications, especially for living cells. Cellular damage can be alleviated by changing imaging parameters to reduce light exposure, often at the expense of image quality. Machine/deep learning methods for single-image super-resolution (SISR) can be applied to restore image quality by upscaling lower-resolution (LR) images to produce high-resolution images (HR). These SISR methods have been successfully applied to photo-realistic images due partly to the abundance of publicly available data. In contrast, the lack of publicly available data partly limits their application and success in scanning confocal microscopy. In this paper, we introduce a large scanning confocal microscopy dataset named SR-CACO-2 that is comprised of low- and high-resolution image pairs marked for three different fluorescent markers. It allows the evaluation of performance of SISR methods on three different upscaling levels (X2, X4, X8). SR-CACO-2 contains the human epithelial cell line Caco-2 (ATCC HTB-37), and it is composed of 22 tiles that have been translated in the form of 9,937 image patches for experiments with SISR methods. Given the new SR-CACO-2 dataset, we also provide benchmarking results for 15 state-of-the-art methods that are representative of the main SISR families. Results show that these methods have limited success in producing high-resolution textures, indicating that SR-CACO-2 represents a challenging problem. Our dataset, code and pretrained weights are available: https://github.com/sbelharbi/sr-caco-2.

A collection of photographic and synthetic images intended for analysis of image processing techniques and quality assessment of displays. Image source: https://testimages.org/

To evaluate the performance on 4K burst images/video, we collect several clips from website. The dataset can be download from : https://drive.google.com/file/d/1YDljUONvyKUO24smTx__CUH_4Zxhle09/view?usp=sharing