This repository lists computer vision datasets grouped by task and uses. The list is in constant update. Let me know if you have any comments or contribution.
| Name | Description | Authors | Article | URL |
|---|---|---|---|---|
| CIFAR10 & CIFAR100 | The CIFAR dataset consists of 60000 32x32 colour images in 10 and 100 classes, with 6000 images per class. There are 50000 training images and 10000 test images. | Alex Krizhevsky; Vinod Nair; Geoffrey Hinton | Learning Multiple Layers of Features from Tiny Images [Report] | https://www.cs.toronto.edu/~kriz/cifar.html |
| Name | Description | Authors | Article | URL |
|---|---|---|---|---|
| MVImageNet | MVImgNet contains 6.5 million frames from 219,188 videos crossing objects from 238 classes. Authors provide an OneDrive link to download the full data. You also require fill a online form to ask for credentials. | Yu, Xianggang; et al | MVImgNet: A Large-scale Dataset of Multi-view Images (CVPR2023) [arXiv] | https://github.com/GAP-LAB-CUHK-SZ/MVImgNet |
| EPFL Multi-view Multi-class Detection dataset | The dataset consists of 23 minutes and 57 seconds of synchronized frames taken at 25fps from 6 different calibrated DV cameras. Also, has a ground truth set that contains 242 annotated multi-view non-consecutive frames. | Gemma Roig; Xavier Boix; Horesh Ben Shitrit; Pascal Fua; et al. | Conditional Random Fields for multi-camera object detection [IEEE] | https://www.epfl.ch/labs/cvlab/data/data-multiclass/ |
| EPFL Multi-View Car Dataset | The dataset on this page contains 20 sequences of cars as they rotate by 360 degrees. There is one image approximately every 3-4 degrees. | M. Özuysal; V. Lepetit; P. Fua | Pose estimation for category specific multiview object localization [IEEE] | https://www.epfl.ch/labs/cvlab/data/data-pose-index-php/ |
| EPFL Multi-camera pedestrians video | This dataset consists of multi-camera sequences used to develop and test a people detection and tracking framework. All sequences feature several synchronized video streams, filming the same area from different angles. The cameras were positioned about 2 meters above the ground. | F. Fleuret; J. Berclaz; R. Lengagne; P. Fua | Multicamera People Tracking with a Probabilistic Occupancy Map [IEEE] | https://www.epfl.ch/labs/cvlab/data/data-pom-index-php/ |