Video

The MNIST database (Modified National Institute of Standards and Technology database) is a large collection of handwritten digits. It has a training set of 60,000 examples, and a test set of 10,000 examples. It is a subset of a larger NIST Special Database 3 (digits written by employees of the United States Census Bureau) and Special Database 1 (digits written by high school students) which contain monochrome images of handwritten digits. The digits have been size-normalized and centered in a fixed-size image. The original black and white (bilevel) images from NIST were size normalized to fit in a 20x20 pixel box while preserving their aspect ratio. The resulting images contain grey levels as a result of the anti-aliasing technique used by the normalization algorithm. the images were centered in a 28x28 image by computing the center of mass of the pixels, and translating the image so as to position this point at the center of the 28x28 field. Source: http://yann.lecun.com/exdb/mnist/ Image Source: https://en.wikipedia.org/wiki/MNIST_database#/media/File:MnistExamples.png

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

UCF101 dataset is an extension of UCF50 and consists of 13,320 video clips, which are classified into 101 categories. These 101 categories can be classified into 5 types (Body motion, Human-human interactions, Human-object interactions, Playing musical instruments and Sports). The total length of these video clips is over 27 hours. All the videos are collected from YouTube and have a fixed frame rate of 25 FPS with the resolution of 320 × 240. Source: Two-stream Collaborative Learning with Spatial-Temporal Attention for Video Classification Image Source: https://www.crcv.ucf.edu/data/UCF101.php

The Kinetics dataset is a large-scale, high-quality dataset for human action recognition in videos. The dataset consists of around 500,000 video clips covering 600 human action classes with at least 600 video clips for each action class. Each video clip lasts around 10 seconds and is labeled with a single action class. The videos are collected from YouTube. Source: Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey

The Flickr30k dataset contains 31,000 images collected from Flickr, together with 5 reference sentences provided by human annotators. Source: Guiding Long-Short Term Memory for Image Caption Generation Image Source: Dual-Path Convolutional Image-Text Embedding with Instance Loss

The HMDB51 dataset is a large collection of realistic videos from various sources, including movies and web videos. The dataset is composed of 6,766 video clips from 51 action categories (such as “jump”, “kiss” and “laugh”), with each category containing at least 101 clips. The original evaluation scheme uses three different training/testing splits. In each split, each action class has 70 clips for training and 30 clips for testing. The average accuracy over these three splits is used to measure the final performance. Source: Action Recognition with Trajectory-Pooled Deep-Convolutional Descriptors Image Source: https://serre-lab.clps.brown.edu/resource/hmdb-a-large-human-motion-database

The ActivityNet dataset contains 200 different types of activities and a total of 849 hours of videos collected from YouTube. ActivityNet is the largest benchmark for temporal activity detection to date in terms of both the number of activity categories and number of videos, making the task particularly challenging. Version 1.3 of the dataset contains 19994 untrimmed videos in total and is divided into three disjoint subsets, training, validation, and testing by a ratio of 2:1:1. On average, each activity category has 137 untrimmed videos. Each video on average has 1.41 activities which are annotated with temporal boundaries. The ground-truth annotations of test videos are not public. Source: Dynamic Temporal Pyramid Network: A Closer Look at Multi-Scale Modeling for Activity Detection

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 Densely Annotation Video Segmentation dataset (DAVIS) is a high quality and high resolution densely annotated video segmentation dataset under two resolutions, 480p and 1080p. There are 50 video sequences with 3455 densely annotated frames in pixel level. 30 videos with 2079 frames are for training and 20 videos with 1376 frames are for validation. Source: TENet: Triple Excitation Network for Video Salient Object Detection

The dataset contains 400 human action classes, with at least 400 video clips for each action. Each clip lasts around 10s and is taken from a different YouTube video. The actions are human focussed and cover a broad range of classes including human-object interactions such as playing instruments, as well as human-human interactions such as shaking hands. Source: https://arxiv.org/abs/1705.06950 Image source: https://arxiv.org/abs/1705.06950

MSR-VTT (Microsoft Research Video to Text) is a large-scale dataset for the open domain video captioning, which consists of 10,000 video clips from 20 categories, and each video clip is annotated with 20 English sentences by Amazon Mechanical Turks. There are about 29,000 unique words in all captions. The standard splits uses 6,513 clips for training, 497 clips for validation, and 2,990 clips for testing. Source: Learning to Discretely Compose Reasoning Module Networksfor Video Captioning

The MPII Human Pose Dataset for single person pose estimation is composed of about 25K images of which 15K are training samples, 3K are validation samples and 7K are testing samples (which labels are withheld by the authors). The images are taken from YouTube videos covering 410 different human activities and the poses are manually annotated with up to 16 body joints. Source: 2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning Image Source: http://human-pose.mpi-inf.mpg.de/

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

NTU RGB+D is a large-scale dataset for RGB-D human action recognition. It involves 56,880 samples of 60 action classes collected from 40 subjects. The actions can be generally divided into three categories: 40 daily actions (e.g., drinking, eating, reading), nine health-related actions (e.g., sneezing, staggering, falling down), and 11 mutual actions (e.g., punching, kicking, hugging). These actions take place under 17 different scene conditions corresponding to 17 video sequences (i.e., S001–S017). The actions were captured using three cameras with different horizontal imaging viewpoints, namely, −45∘,0∘, and +45∘. Multi-modality information is provided for action characterization, including depth maps, 3D skeleton joint position, RGB frames, and infrared sequences. The performance evaluation is performed by a cross-subject test that split the 40 subjects into training and test groups, and by a cross-view test that employed one camera (+45∘) for testing, and the other two cameras for training. Source: Action Recognition for Depth Video using Multi-view Dynamic Images

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.

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

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

The Microsoft Research Video Description Corpus (MSVD) dataset consists of about 120K sentences collected during the summer of 2010. Workers on Mechanical Turk were paid to watch a short video snippet and then summarize the action in a single sentence. The result is a set of roughly parallel descriptions of more than 2,000 video snippets. Because the workers were urged to complete the task in the language of their choice, both paraphrase and bilingual alternations are captured in the data. Source: https://www.microsoft.com/en-us/download/details.aspx?id=52422&from=https%3A%2F%2Fresearch.microsoft.com%2Fen-us%2Fdownloads%2F38cf15fd-b8df-477e-a4e4-a4680caa75af%2F Image Source: https://arxiv.org/pdf/1609.06782.pdf

The THUMOS14 (THUMOS 2014) dataset is a large-scale video dataset that includes 1,010 videos for validation and 1,574 videos for testing from 20 classes. Among all the videos, there are 220 and 212 videos with temporal annotations in validation and testing set, respectively. Source: Learning to Localize Actions from Moments Image Source: http://crcv.ucf.edu/THUMOS14/

DAVIS17 is a dataset for video object segmentation. It contains a total of 150 videos - 60 for training, 30 for validation, 60 for testing Source: Siam R-CNN: Visual Tracking by Re-Detection Image Source: https://www.researchgate.net/figure/LucidTracker-qualitative-results-on-DAVIS-17-test-dev-set-Frames-sampled-along-the_fig5_331792902

The 20BN-SOMETHING-SOMETHING V2 dataset is a large collection of labeled video clips that show humans performing pre-defined basic actions with everyday objects. The dataset was created by a large number of crowd workers. It allows machine learning models to develop fine-grained understanding of basic actions that occur in the physical world. It contains 220,847 videos, with 168,913 in the training set, 24,777 in the validation set and 27,157 in the test set. There are 174 labels. Source Image Source

HowTo100M is a large-scale dataset of narrated videos with an emphasis on instructional videos where content creators teach complex tasks with an explicit intention of explaining the visual content on screen. HowTo100M features a total of: - 136M video clips with captions sourced from 1.2M Youtube videos (15 years of video) - 23k activities from domains such as cooking, hand crafting, personal care, gardening or fitness Each video is associated with a narration available as subtitles automatically downloaded from Youtube. Source: HowTo100M

The efforts to create a non-trivial and publicly available dataset for action recognition was initiated at the KTH Royal Institute of Technology in 2004. The KTH dataset is one of the most standard datasets, which contains six actions: walk, jog, run, box, hand-wave, and hand clap. To account for performance nuance, each action is performed by 25 different individuals, and the setting is systematically altered for each action per actor. Setting variations include: outdoor (s1), outdoor with scale variation (s2), outdoor with different clothes (s3), and indoor (s4). These variations test the ability of each algorithm to identify actions independent of the background, appearance of the actors, and the scale of the actors. Source: Review of Action Recognition and Detection Methods

The Shanghaitech dataset is a large-scale crowd counting dataset. It consists of 1198 annotated crowd images. The dataset is divided into two parts, Part-A containing 482 images and Part-B containing 716 images. Part-A is split into train and test subsets consisting of 300 and 182 images, respectively. Part-B is split into train and test subsets consisting of 400 and 316 images. Each person in a crowd image is annotated with one point close to the center of the head. In total, the dataset consists of 330,165 annotated people. Images from Part-A were collected from the Internet, while images from Part-B were collected on the busy streets of Shanghai. Source: Iterative Crowd Counting Image Source: Li et al

WebVid contains 10 million video clips with captions, sourced from the web. The videos are diverse and rich in their content. Both the full 10M set and a 2.5M subset is available for download: https://github.com/m-bain/webvid-dataset

The ActivityNet Captions dataset is built on ActivityNet v1.3 which includes 20k YouTube untrimmed videos with 100k caption annotations. The videos are 120 seconds long on average. Most of the videos contain over 3 annotated events with corresponding start/end time and human-written sentences, which contain 13.5 words on average. The number of videos in train/validation/test split is 10024/4926/5044, respectively. Source: Bidirectional Attentive Fusion with Context Gating for Dense Video Captioning Image Source: https://cs.stanford.edu/people/ranjaykrishna/densevid/

JHMDB is an action recognition dataset that consists of 960 video sequences belonging to 21 actions. It is a subset of the larger HMDB51 dataset collected from digitized movies and YouTube videos. The dataset contains video and annotation for puppet flow per frame (approximated optimal flow on the person), puppet mask per frame, joint positions per frame, action label per clip and meta label per clip (camera motion, visible body parts, camera viewpoint, number of people, video quality). Source: Unsupervised Deep Metric Learning via Orthogonality based Probabilistic Loss Image Source: https://arxiv.org/pdf/1712.06316.pdf

The GOT-10k dataset contains more than 10,000 video segments of real-world moving objects and over 1.5 million manually labelled bounding boxes. The dataset contains more than 560 classes of real-world moving objects and 80+ classes of motion patterns. Source: http://got-10k.aitestunion.com/ Image Source: https://arxiv.org/pdf/1810.11981.pdf

Charades-STA is a new dataset built on top of Charades by adding sentence temporal annotations. Source: TALL: Temporal Activity Localization via Language Query

DAVIS16 is a dataset for video object segmentation which consists of 50 videos in total (30 videos for training and 20 for testing). Per-frame pixel-wise annotations are offered. Source: Learning Discriminative Feature with CRF for Unsupervised Video Object Segmentation Image Source: https://davischallenge.org/

The Middlebury Stereo dataset consists of high-resolution stereo sequences with complex geometry and pixel-accurate ground-truth disparity data. The ground-truth disparities are acquired using a novel technique that employs structured lighting and does not require the calibration of the light projectors. Source: https://vision.middlebury.edu/stereo/data/ Image Source: https://www.researchgate.net/figure/The-stereo-matching-results-on-the-Middlebury-dataset-From-left-to-right-each-set-of_fig3_273399625

The Vimeo-90K is a large-scale high-quality video dataset for lower-level video processing. It proposes three different video processing tasks: frame interpolation, video denoising/deblocking, and video super-resolution. Source: https://arxiv.org/pdf/1711.09078.pdf Image Source: http://toflow.csail.mit.edu/

The Distinct Describable Moments (DiDeMo) dataset is one of the largest and most diverse datasets for the temporal localization of events in videos given natural language descriptions. The videos are collected from Flickr and each video is trimmed to a maximum of 30 seconds. The videos in the dataset are divided into 5-second segments to reduce the complexity of annotation. The dataset is split into training, validation and test sets containing 8,395, 1,065 and 1,004 videos respectively. The dataset contains a total of 26,892 moments and one moment could be associated with descriptions from multiple annotators. The descriptions in DiDeMo dataset are detailed and contain camera movement, temporal transition indicators, and activities. Moreover, the descriptions in DiDeMo are verified so that each description refers to a single moment. Source: Weakly Supervised Video Moment Retrieval From Text Queries Image Source: https://www.di.ens.fr/~miech/datasetviz/

The ShanghaiTech Campus dataset has 13 scenes with complex light conditions and camera angles. It contains 130 abnormal events and over 270, 000 training frames. Moreover, both the frame-level and pixel-level ground truth of abnormal events are annotated in this dataset.

Youtube-VOS is a Video Object Segmentation dataset that contains 4,453 videos - 3,471 for training, 474 for validation, and 508 for testing. The training and validation videos have pixel-level ground truth annotations for every 5th frame (6 fps). It also contains Instance Segmentation annotations. It has more than 7,800 unique objects, 190k high-quality manual annotations and more than 340 minutes in duration. Source: CapsuleVOS: Semi-Supervised Video Object Segmentation Using Capsule Routing Image Source: https://youtube-vos.org/