2D Classification

The ImageNet dataset contains 14,197,122 annotated images according to the WordNet hierarchy. Since 2010 the dataset is used in the ImageNet Large Scale Visual Recognition Challenge (ILSVRC), a benchmark in image classification and object detection. The publicly released dataset contains a set of manually annotated training images. A set of test images is also released, with the manual annotations withheld. ILSVRC annotations fall into one of two categories: (1) image-level annotation of a binary label for the presence or absence of an object class in the image, e.g., “there are cars in this image” but “there are no tigers,” and (2) object-level annotation of a tight bounding box and class label around an object instance in the image, e.g., “there is a screwdriver centered at position (20,25) with width of 50 pixels and height of 30 pixels”. The ImageNet project does not own the copyright of the images, therefore only thumbnails and URLs of images are provided. * Total number of non-empty WordNet synsets: 21841 * Total number of images: 14197122 * Number of images with bounding box annotations: 1,034,908 * Number of synsets with SIFT features: 1000 * Number of images with SIFT features: 1.2 million Source: ImageNet Large Scale Visual Recognition Challenge Image Source: https://cs.stanford.edu/people/karpathy/cnnembed/

The COCO (Common Objects in Context) dataset is a large-scale object detection, segmentation, and captioning dataset. It is designed to encourage research on a wide variety of object categories and is commonly used for benchmarking computer vision models. It is an essential dataset for researchers and developers working on object detection, segmentation, and pose estimation tasks.

Cityscapes is a large-scale database which focuses on semantic understanding of urban street scenes. It provides semantic, instance-wise, and dense pixel annotations for 30 classes grouped into 8 categories (flat surfaces, humans, vehicles, constructions, objects, nature, sky, and void). The dataset consists of around 5000 fine annotated images and 20000 coarse annotated ones. Data was captured in 50 cities during several months, daytimes, and good weather conditions. It was originally recorded as video so the frames were manually selected to have the following features: large number of dynamic objects, varying scene layout, and varying background. Source: A Review on Deep Learning Techniques Applied to Semantic Segmentation Image Source: https://www.cityscapes-dataset.com/dataset-overview/

KITTI (Karlsruhe Institute of Technology and Toyota Technological Institute) is one of the most popular datasets for use in mobile robotics and autonomous driving. It consists of hours of traffic scenarios recorded with a variety of sensor modalities, including high-resolution RGB, grayscale stereo cameras, and a 3D laser scanner. Despite its popularity, the dataset itself does not contain ground truth for semantic segmentation. However, various researchers have manually annotated parts of the dataset to fit their necessities. Álvarez et al. generated ground truth for 323 images from the road detection challenge with three classes: road, vertical, and sky. Zhang et al. annotated 252 (140 for training and 112 for testing) acquisitions – RGB and Velodyne scans – from the tracking challenge for ten object categories: building, sky, road, vegetation, sidewalk, car, pedestrian, cyclist, sign/pole, and fence. Ros et al. labeled 170 training images and 46 testing images (from the visual odometry challenge) with 11 classes: building, tree, sky, car, sign, road, pedestrian, fence, pole, sidewalk, and bicyclist. Source: A Review on Deep Learning Techniques Applied to Semantic Segmentation Image Source: http://www.cvlibs.net/datasets/kitti/eval_object.php?obj_benchmark=3d

The LibriSpeech corpus is a collection of approximately 1,000 hours of audiobooks that are a part of the LibriVox project. Most of the audiobooks come from the Project Gutenberg. The training data is split into 3 partitions of 100hr, 360hr, and 500hr sets while the dev and test data are split into the ’clean’ and ’other’ categories, respectively, depending upon how well or challenging Automatic Speech Recognition systems would perform against. Each of the dev and test sets is around 5hr in audio length. This corpus also provides the n-gram language models and the corresponding texts excerpted from the Project Gutenberg books, which contain 803M tokens and 977K unique words. Source: State-of-the-art Speech Recognition using Multi-stream Self-attention with Dilated 1D Convolutions

The nuScenes dataset is a large-scale autonomous driving dataset. The dataset has 3D bounding boxes for 1000 scenes collected in Boston and Singapore. Each scene is 20 seconds long and annotated at 2Hz. This results in a total of 28130 samples for training, 6019 samples for validation and 6008 samples for testing. The dataset has the full autonomous vehicle data suite: 32-beam LiDAR, 6 cameras and radars with complete 360° coverage. The 3D object detection challenge evaluates the performance on 10 classes: cars, trucks, buses, trailers, construction vehicles, pedestrians, motorcycles, bicycles, traffic cones and barriers. Source: PointPainting: Sequential Fusion for 3D Object Detection

ScanNet is an instance-level indoor RGB-D dataset that includes both 2D and 3D data. It is a collection of labeled voxels rather than points or objects. Up to now, ScanNet v2, the newest version of ScanNet, has collected 1513 annotated scans with an approximate 90% surface coverage. In the semantic segmentation task, this dataset is marked in 20 classes of annotated 3D voxelized objects. Source: A Review of Point Cloud Semantic Segmentation Image Source: http://www.scan-net.org/

Visual Genome contains Visual Question Answering data in a multi-choice setting. It consists of 101,174 images from MSCOCO with 1.7 million QA pairs, 17 questions per image on average. Compared to the Visual Question Answering dataset, Visual Genome represents a more balanced distribution over 6 question types: What, Where, When, Who, Why and How. The Visual Genome dataset also presents 108K images with densely annotated objects, attributes and relationships. Source: RaAM: A Relation-aware Attention Model for Visual Question Answering Image Source: Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations

The NYU-Depth V2 data set is comprised of video sequences from a variety of indoor scenes as recorded by both the RGB and Depth cameras from the Microsoft Kinect. It features: * 1449 densely labeled pairs of aligned RGB and depth images * 464 new scenes taken from 3 cities * 407,024 new unlabeled frames * Each object is labeled with a class and an instance number. The dataset has several components: * Labeled: A subset of the video data accompanied by dense multi-class labels. This data has also been preprocessed to fill in missing depth labels. * Raw: The raw RGB, depth and accelerometer data as provided by the Kinect. * Toolbox: Useful functions for manipulating the data and labels. Source: https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html Image Source: https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html

The Human3.6M dataset is one of the largest motion capture datasets, which consists of 3.6 million human poses and corresponding images captured by a high-speed motion capture system. There are 4 high-resolution progressive scan cameras to acquire video data at 50 Hz. The dataset contains activities by 11 professional actors in 17 scenarios: discussion, smoking, taking photo, talking on the phone, etc., as well as provides accurate 3D joint positions and high-resolution videos. Source: Space-Time Representation of People Based on 3D Skeletal Data: A Review Image Source: Yu et al

The GQA dataset is a large-scale visual question answering dataset with real images from the Visual Genome dataset and balanced question-answer pairs. Each training and validation image is also associated with scene graph annotations describing the classes and attributes of those objects in the scene, and their pairwise relations. Along with the images and question-answer pairs, the GQA dataset provides two types of pre-extracted visual features for each image – convolutional grid features of size 7×7×2048 extracted from a ResNet-101 network trained on ImageNet, and object detection features of size Ndet×2048 (where Ndet is the number of detected objects in each image with a maximum of 100 per image) from a Faster R-CNN detector. Source: Language-Conditioned Graph Networks for Relational Reasoning Image Source: https://arxiv.org/pdf/1902.09506.pdf

LVIS is a dataset for long tail instance segmentation. It has annotations for over 1000 object categories in 164k images. Source: LVIS

The Stanford 3D Indoor Scene Dataset (S3DIS) dataset contains 6 large-scale indoor areas with 271 rooms. Each point in the scene point cloud is annotated with one of the 13 semantic categories. Source: Grid-GCN for Fast and Scalable Point Cloud Learning Image Source: https://www.researchgate.net/figure/Examples-of-classified-scenes-in-S3DIS-dataset-left-with-groundtruth-right_fig2_328307943

The Waymo Open Dataset is comprised of high resolution sensor data collected by autonomous vehicles operated by the Waymo Driver in a wide variety of conditions. The Waymo Open Dataset currently contains 1,950 segments. The authors plan to grow this dataset in the future. Currently the datasets includes: * 1,950 segments of 20s each, collected at 10Hz (390,000 frames) in diverse geographies and conditions * Sensor data * 1 mid-range lidar * 4 short-range lidars * 5 cameras (front and sides) * Synchronized lidar and camera data * Lidar to camera projections * Sensor calibrations and vehicle poses * Labeled data * Labels for 4 object classes - Vehicles, Pedestrians, Cyclists, Signs * High-quality labels for lidar data in 1,200 segments * 12.6M 3D bounding box labels with tracking IDs on lidar data * High-quality labels for camera data in 1,000 segments * 11.8M 2D bounding box labels with tracking IDs on camera data

The SUN RGBD dataset contains 10335 real RGB-D images of room scenes. Each RGB image has a corresponding depth and segmentation map. As many as 700 object categories are labeled. The training and testing sets contain 5285 and 5050 images, respectively. Source: Mix and match networks: multi-domain alignment for unpaired image-to-image translation Image Source: https://rgbd.cs.princeton.edu/

This CSTR VCTK Corpus includes speech data uttered by 110 English speakers with various accents. Each speaker reads out about 400 sentences, which were selected from a newspaper, the rainbow passage and an elicitation paragraph used for the speech accent archive. The newspaper texts were taken from Herald Glasgow, with permission from Herald & Times Group. Each speaker has a different set of the newspaper texts selected based a greedy algorithm that increases the contextual and phonetic coverage. The details of the text selection algorithms are described in the following paper: C. Veaux, J. Yamagishi and S. King, "The voice bank corpus: Design, collection and data analysis of a large regional accent speech database," https://doi.org/10.1109/ICSDA.2013.6709856. The rainbow passage and elicitation paragraph are the same for all speakers. The rainbow passage can be found at International Dialects of English Archive: (http://web.ku.edu/~idea/readings/rainbow.htm). The elicitation paragraph is identical to the one used for the speech accent archive (http://accent.gmu.edu). The details of the the speech accent archive can be found at http://www.ualberta.ca/~aacl2009/PDFs/WeinbergerKunath2009AACL.pdf. All speech data was recorded using an identical recording setup: an omni-directional microphone (DPA 4035) and a small diaphragm condenser microphone with very wide bandwidth (Sennheiser MKH 800), 96kHz sampling frequency at 24 bits and in a hemi-anechoic chamber of the University of Edinburgh. (However, two speakers, p280 and p315 had technical issues of the audio recordings using MKH 800). All recordings were converted into 16 bits, were downsampled to 48 kHz, and were manually end-pointed. Source: CSTR VCTK Corpus: English Multi-speaker Corpus for CSTR Voice Cloning Toolkit (version 0.92)

Datasets drive vision progress, yet existing driving datasets are impoverished in terms of visual content and supported tasks to study multitask learning for autonomous driving. Researchers are usually constrained to study a small set of problems on one dataset, while real-world computer vision applications require performing tasks of various complexities. We construct BDD100K, the largest driving video dataset with 100K videos and 10 tasks to evaluate the exciting progress of image recognition algorithms on autonomous driving. The dataset possesses geographic, environmental, and weather diversity, which is useful for training models that are less likely to be surprised by new conditions. Based on this diverse dataset, we build a benchmark for heterogeneous multitask learning and study how to solve the tasks together. Our experiments show that special training strategies are needed for existing models to perform such heterogeneous tasks. BDD100K opens the door for future studies in this important venue. More detail is at the dataset home page.

Common Voice is an audio dataset that consists of a unique MP3 and corresponding text file. There are 9,283 recorded hours in the dataset. The dataset also includes demographic metadata like age, sex, and accent. The dataset consists of 7,335 validated hours in 60 languages.

The Charades dataset is composed of 9,848 videos of daily indoors activities with an average length of 30 seconds, involving interactions with 46 objects classes in 15 types of indoor scenes and containing a vocabulary of 30 verbs leading to 157 action classes. Each video in this dataset is annotated by multiple free-text descriptions, action labels, action intervals and classes of interacting objects. 267 different users were presented with a sentence, which includes objects and actions from a fixed vocabulary, and they recorded a video acting out the sentence. In total, the dataset contains 66,500 temporal annotations for 157 action classes, 41,104 labels for 46 object classes, and 27,847 textual descriptions of the videos. In the standard split there are7,986 training video and 1,863 validation video. Source: Temporal Reasoning Graph for Activity Recognition

MVTec AD is a dataset for benchmarking anomaly detection methods with a focus on industrial inspection. It contains over 5000 high-resolution images divided into fifteen different object and texture categories. Each category comprises a set of defect-free training images and a test set of images with various kinds of defects as well as images without defects. There are two common metrics: Detection AUROC and Segmentation (or pixelwise) AUROC Detection (or, classification) methods output single float (anomaly score) per input test image. Segmentation methods output anomaly probability for each pixel. "To assess segmentation performance, we evaluate the relative per-region overlap of the segmentation with the ground truth. To get an additional performance measure that is independent of the determined threshold, we compute the area under the receiver operating characteristic curve (ROC AUC). We define the true positive rate as the percentage of pixels that were correctly classified as anomalous" [1] Later segmentation metric was improved to balance regions with small and large area, see PRO-AUC and other in [2] Source: MVTEC ANOMALY DETECTION DATASET Image Source: https://www.mvtec.com/company/research/datasets/mvtec-ad/ [1] Paul Bergmann et al, "MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection" [2] Bergmann, P., Batzner, K., Fauser, M. et al. The MVTec Anomaly Detection Dataset: A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection. Int J Comput Vis (2021). https://doi.org/10.1007/s11263-020-01400-4

The GoPro dataset for deblurring consists of 3,214 blurred images with the size of 1,280×720 that are divided into 2,103 training images and 1,111 test images. The dataset consists of pairs of a realistic blurry image and the corresponding ground truth sharp image that are obtained by a high-speed camera. Source: Down-Scaling with Learned Kernels in Multi-Scale Deep Neural Networksfor Non-Uniform Single Image Deblurring Image Source: Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring

Argoverse is a tracking benchmark with over 30K scenarios collected in Pittsburgh and Miami. Each scenario is a sequence of frames sampled at 10 HZ. Each sequence has an interesting object called “agent”, and the task is to predict the future locations of agents in a 3 seconds future horizon. The sequences are split into training, validation and test sets, which have 205,942, 39,472 and 78,143 sequences respectively. These splits have no geographical overlap. Source: Learning Lane Graph Representations for Motion Forecasting Image Source: https://arxiv.org/pdf/1911.02620.pdf

The DukeMTMC-reID (Duke Multi-Tracking Multi-Camera ReIDentification) dataset is a subset of the DukeMTMC for image-based person re-ID. The dataset is created from high-resolution videos from 8 different cameras. It is one of the largest pedestrian image datasets wherein images are cropped by hand-drawn bounding boxes. The dataset consists 16,522 training images of 702 identities, 2,228 query images of the other 702 identities and 17,661 gallery images. NOTE: This dataset has been retracted. Source: Deep Co-attention based Comparators For Relative Representation Learning in Person Re-identification

Manga109 has been compiled by the Aizawa Yamasaki Matsui Laboratory, Department of Information and Communication Engineering, the Graduate School of Information Science and Technology, the University of Tokyo. The compilation is intended for use in academic research on the media processing of Japanese manga. Manga109 is composed of 109 manga volumes drawn by professional manga artists in Japan. These manga were commercially made available to the public between the 1970s and 2010s, and encompass a wide range of target readerships and genres (see the table in Explore for further details.) Most of the manga in the compilation are available at the manga library “Manga Library Z” (formerly the “Zeppan Manga Toshokan” library of out-of-print manga). Source: Manga109 Image Source: https://arxiv.org/pdf/1510.04389v1.pdf

DOTA is a large-scale dataset for object detection in aerial images. It can be used to develop and evaluate object detectors in aerial images. The images are collected from different sensors and platforms. Each image is of the size in the range from 800 × 800 to 20,000 × 20,000 pixels and contains objects exhibiting a wide variety of scales, orientations, and shapes. The instances in DOTA images are annotated by experts in aerial image interpretation by arbitrary (8 d.o.f.) quadrilateral. We will continue to update DOTA, to grow in size and scope to reflect evolving real-world conditions. Now it has three versions: DOTA-v1.0 contains 15 common categories, 2,806 images and 188, 282 instances. The proportions of the training set, validation set, and testing set in DOTA-v1.0 are 1/2, 1/6, and 1/3, respectively. DOTA-v1.5 uses the same images as DOTA-v1.0, but the extremely small instances (less than 10 pixels) are also annotated. Moreover, a new category, ”container crane” is added. It contains 403,318 instances in total. The number of images and dataset splits are the same as DOTA-v1.0. This version was released for the DOAI Challenge 2019 on Object Detection in Aerial Images in conjunction with IEEE CVPR 2019. DOTA-v2.0 collects more Google Earth, GF-2 Satellite, and aerial images. There are 18 common categories, 11,268 images and 1,793,658 instances in DOTA-v2.0. Compared to DOTA-v1.5, it further adds the new categories of ”airport” and ”helipad”. The 11,268 images of DOTA are split into training, validation, test-dev, and test-challenge sets. To avoid the problem of overfitting, the proportion of training and validation set is smaller than the test set. Furthermore, we have two test sets, namely test-dev and test-challenge. Training contains 1,830 images and 268,627 instances. Validation contains 593 images and 81,048 instances. We released the images and ground truths for training and validation sets. Test-dev contains 2,792 images and 353,346 instances. We released the images but not the ground truths. Test-challenge contains 6,053 images and 1,090,637 instances. Source: https://captain-whu.github.io/DOTA/index.html Image Source: https://captain-whu.github.io/DOTA/

DUTS is a saliency detection dataset containing 10,553 training images and 5,019 test images. All training images are collected from the ImageNet DET training/val sets, while test images are collected from the ImageNet DET test set and the SUN data set. Both the training and test set contain very challenging scenarios for saliency detection. Accurate pixel-level ground truths are manually annotated by 50 subjects. Source: http://saliencydetection.net/duts/ Image Source: https://ieeexplore.ieee.org/document/8099887

PASCAL-S is a dataset for salient object detection consisting of a set of 850 images from PASCAL VOC 2010 validation set with multiple salient objects on the scenes. Source: Structured Modeling of Joint Deep Feature and Prediction Refinement for Salient Object Detection

Description: 10,000 People - Human Pose Recognition Data. This dataset includes indoor and outdoor scenes.This dataset covers males and females. Age distribution ranges from teenager to the elderly, the middle-aged and young people are the majorities. The data diversity includes different shooting heights, different ages, different light conditions, different collecting environment, clothes in different seasons, multiple human poses. For each subject, the labels of gender, race, age, collecting environment and clothes were annotated. The data can be used for human pose recognition and other tasks. Data size: 10,000 people Race distribution: Asian (Chinese)

Foggy Cityscapes is a synthetic foggy dataset which simulates fog on real scenes. Each foggy image is rendered with a clear image and depth map from Cityscapes. Thus the annotations and data split in Foggy Cityscapes are inherited from Cityscapes. Source: Exploring Object Relation in Mean Teacher for Cross-Domain Detection Image Source: http://people.ee.ethz.ch/~csakarid/SFSU_synthetic/

KITTI-360 is a large-scale dataset that contains rich sensory information and full annotations. It is the successor of the popular KITTI dataset, providing more comprehensive semantic/instance labels in 2D and 3D, richer 360 degree sensory information (fisheye images and pushbroom laser scans), very accurate and geo-localized vehicle and camera poses, and a series of new challenging benchmarks.

The Pascal3D+ multi-view dataset consists of images in the wild, i.e., images of object categories exhibiting high variability, captured under uncontrolled settings, in cluttered scenes and under many different poses. Pascal3D+ contains 12 categories of rigid objects selected from the PASCAL VOC 2012 dataset. These objects are annotated with pose information (azimuth, elevation and distance to camera). Pascal3D+ also adds pose annotated images of these 12 categories from the ImageNet dataset. Source: Convolutional Models for Joint Object Categorization and Pose Estimation Image Source: Beyond PASCAL: A benchmark for 3D object detection in the wild

HKU-IS is a visual saliency prediction dataset which contains 4447 challenging images, most of which have either low contrast or multiple salient objects. Source: Deep Contrast Learning for Salient Object Detection Image Source: https://sites.google.com/site/ligb86/mdfsaliency/

The DUT-OMRON dataset is used for evaluation of Salient Object Detection task and it contains 5,168 high quality images. The images have one or more salient objects and relatively cluttered background. Source: Global Context-Aware Progressive Aggregation Network for Salient Object Detection

The 300-W is a face dataset that consists of 300 Indoor and 300 Outdoor in-the-wild images. It covers a large variation of identity, expression, illumination conditions, pose, occlusion and face size. The images were downloaded from google.com by making queries such as “party”, “conference”, “protests”, “football” and “celebrities”. Compared to the rest of in-the-wild datasets, the 300-W database contains a larger percentage of partially-occluded images and covers more expressions than the common “neutral” or “smile”, such as “surprise” or “scream”. Images were annotated with the 68-point mark-up using a semi-automatic methodology. The images of the database were carefully selected so that they represent a characteristic sample of challenging but natural face instances under totally unconstrained conditions. Thus, methods that achieve accurate performance on the 300-W database can demonstrate the same accuracy in most realistic cases. Many images of the database contain more than one annotated faces (293 images with 1 face, 53 images with 2 faces and 53 images with [3, 7] faces). Consequently, the database consists of 600 annotated face instances, but 399 unique images. Finally, there is a large variety of face sizes. Specifically, 49.3% of the faces have size in the range [48.6k, 2.0M] and the overall mean size is 85k (about 292 × 292) pixels. Source: https://ibug.doc.ic.ac.uk/media/uploads/documents/sagonas_2016_imavis.pdf Image Source: https://www.researchgate.net/profile/Xuanyi_Dong/publication/323722412/figure/fig1/AS:679426136227845@1538999222829/Face-samples-from-300-W-dataset-Different-faces-have-different-styles-whereas-the-style_Q640.jpg

Kvasir-SEG is an open-access dataset of gastrointestinal polyp images and corresponding segmentation masks, manually annotated by a medical doctor and then verified by an experienced gastroenterologist. Source: Kvasir-SEG: A Segmented Polyp Dataset Image Source: https://datasets.simula.no/kvasir-seg/