The FLIC dataset contains 5003 images from popular Hollywood movies. The images were obtained by running a state-of-the-art person detector on every tenth frame of 30 movies. People detected with high confidence (roughly 20K candidates) were then sent to the crowdsourcing marketplace Amazon Mechanical Turk to obtain ground truth labelling. Each image was annotated by five Turkers to label 10 upper body joints. The median-of-five labelling was taken in each image to be robust to outlier annotation. Finally, images were rejected manually by if the person was occluded or severely non-frontal.
45 PAPERS • 2 BENCHMARKS
A novel large-scale corpus of manual annotations for the SoccerNet video dataset, along with open challenges to encourage more research in soccer understanding and broadcast production.
45 PAPERS • 7 BENCHMARKS
DailyActivity3D dataset is a daily activity dataset captured by a Kinect device. There are 16 activity types: drink, eat, read book, call cellphone, write on a paper, use laptop, use vacuum cleaner, cheer up, sit still, toss paper, play game, lay down on sofa, walk, play guitar, stand up, sit down. If possible, each subject performs an activity in two different poses: “sitting on sofa” and “standing”. The total number of the activity samples is 320. This dataset is designed to cover human’s daily activities in the living room. When the performer stands close to the sofa or sits on the sofa, the 3D joint positions extracted by the skeleton tracker are very noisy. Moreover, most of the activities involve the humans-object interactions. Thus this dataset is more challenging.
44 PAPERS • 1 BENCHMARK
Over a period of three years (2009 - 2011) the daily news and weather forecast airings of the German public tv-station PHOENIX featuring sign language interpretation have been recorded and the weather forecasts of a subset of 386 editions have been transcribed using gloss notation. Furthermore, we used automatic speech recognition with manual cleaning to transcribe the original German speech. As such, this corpus allows to train end-to-end sign language translation systems from sign language video input to spoken language.
44 PAPERS • 2 BENCHMARKS
This data set was prepared from 88 open-source YouTube cooking videos. The YouCook dataset contains videos of people cooking various recipes. The videos were downloaded from YouTube and are all in the third-person viewpoint; they represent a significantly more challenging visual problem than existing cooking and kitchen datasets (the background kitchen/scene is different for many and most videos have dynamic camera changes). In addition, frame-by-frame object and action annotations are provided for training data (as well as a number of precomputed low-level features). Finally, each video has a number of human provided natural language descriptions (on average, there are eight different descriptions per video). This dataset has been created to serve as a benchmark in describing complex real-world videos with natural language descriptions.
44 PAPERS • NO BENCHMARKS YET
3,859 high-resolution YouTube videos, 2,985 training videos, 421 validation videos and 453 test videos. An improved 40-category label set by merging eagle and owl into bird, ape into monkey, deleting hands, and adding flying disc, squirrel and whale 8,171 unique video instances 232k high-quality manual annotations
CityFlow is a city-scale traffic camera dataset consisting of more than 3 hours of synchronized HD videos from 40 cameras across 10 intersections, with the longest distance between two simultaneous cameras being 2.5 km. The dataset contains more than 200K annotated bounding boxes covering a wide range of scenes, viewing angles, vehicle models, and urban traffic flow conditions.
43 PAPERS • 2 BENCHMARKS
Room-Across-Room (RxR) is a multilingual dataset for Vision-and-Language Navigation (VLN) for Matterport3D environments. In contrast to related datasets such as Room-to-Room (R2R), RxR is 10x larger, multilingual (English, Hindi and Telugu), with longer and more variable paths, and it includes and fine-grained visual groundings that relate each word to pixels/surfaces in the environment.
43 PAPERS • 1 BENCHMARK
MovieNet is a holistic dataset for movie understanding. MovieNet contains 1,100 movies with a large amount of multi-modal data, e.g. trailers, photos, plot descriptions, etc.. Besides, different aspects of manual annotations are provided in MovieNet, including 1.1M characters with bounding boxes and identities, 42K scene boundaries, 2.5K aligned description sentences, 65K tags of place and action, and 92 K tags of cinematic style.
42 PAPERS • NO BENCHMARKS YET
A2D (Actor-Action Dataset) is a dataset for simultaneously inferring actors and actions in videos. A2D has seven actor classes (adult, baby, ball, bird, car, cat, and dog) and eight action classes (climb, crawl, eat, fly, jump, roll, run, and walk) not including the no-action class, which we also consider. The A2D has 3,782 videos with at least 99 instances per valid actor-action tuple and videos are labeled with both pixel-level actors and actions for sampled frames. The A2D dataset serves as a large-scale testbed for various vision problems: video-level single- and multiple-label actor-action recognition, instance-level object segmentation/co-segmentation, as well as pixel-level actor-action semantic segmentation to name a few.
41 PAPERS • 1 BENCHMARK
Assembly101 is a new procedural activity dataset featuring 4321 videos of people assembling and disassembling 101 "take-apart" toy vehicles. Participants work without fixed instructions, and the sequences feature rich and natural variations in action ordering, mistakes, and corrections. Assembly101 is the first multi-view action dataset, with simultaneous static (8) and egocentric (4) recordings. Sequences are annotated with more than 100K coarse and 1M fine-grained action segments, and 18M 3D hand poses. We benchmark on three action understanding tasks: recognition, anticipation and temporal segmentation. Additionally, we propose a novel task of detecting mistakes. The unique recording format and rich set of annotations allow us to investigate generalization to new toys, cross-view transfer, long-tailed distributions, and pose vs. appearance. We envision that Assembly101 will serve as a new challenge to investigate various activity understanding problems.
38 PAPERS • 4 BENCHMARKS
The dataset contains over 15K images of 20 people (6 females and 14 males - 4 people were recorded twice). For each frame, a depth image, the corresponding rgb image (both 640x480 pixels), and the annotation is provided. The head pose range covers about +-75 degrees yaw and +-60 degrees pitch. Ground truth is provided in the form of the 3D location of the head and its rotation.
38 PAPERS • 1 BENCHMARK
PandaSet is a dataset produced by a complete, high-precision autonomous vehicle sensor kit with a no-cost commercial license. The dataset was collected using one 360x360 mechanical spinning LiDAR, one forward-facing, long-range LiDRAR, and 6 cameras. The datasets contains more than 100 scenes, each of which is 8 seconds long, and provides 28 types of labels for object classification and 37 types of annotations for semantic segmentation.
38 PAPERS • NO BENCHMARKS YET
Inferring human-scene contact (HSC) is the first step toward understanding how humans interact with their surroundings. While detecting 2D human-object interaction (HOI) and reconstructing 3D human pose and shape (HPS) have enjoyed significant progress, reasoning about 3D human-scene contact from a single image is still challenging. Existing HSC detection methods consider only a few types of predefined contact, often reduce body and scene to a small number of primitives, and even overlook image evidence. To predict human-scene contact from a single image, we address the limitations above from both data and algorithmic perspectives. We capture a new dataset called RICH for “Real scenes, Interaction, Contact and Humans.” RICH contains multiview outdoor/indoor video sequences at 4K resolution, ground-truth 3D human bodies captured using markerless motion capture, 3D body scans, and high resolution 3D scene scans. A key feature of RICH is that it also contains accurate vertex-level contact
TAO is a federated dataset for Tracking Any Object, containing 2,907 high resolution videos, captured in diverse environments, which are half a minute long on average. A bottom-up approach was used for discovering a large vocabulary of 833 categories, an order of magnitude more than prior tracking benchmarks.
CSL-Daily (Chinese Sign Language Corpus) is a large-scale continuous SLT dataset. It provides both spoken language translations and gloss-level annotations. The topic revolves around people's daily lives (e.g., travel, shopping, medical care), the most likely SLT application scenario.
37 PAPERS • 4 BENCHMARKS
The EPIC-KITCHENS-55 dataset comprises a set of 432 egocentric videos recorded by 32 participants in their kitchens at 60fps with a head mounted camera. There is no guiding script for the participants who freely perform activities in kitchens related to cooking, food preparation or washing up among others. Each video is split into short action segments (mean duration is 3.7s) with specific start and end times and a verb and noun annotation describing the action (e.g. ‘open fridge‘). The verb classes are 125 and the noun classes 331. The dataset is divided into one train and two test splits.
36 PAPERS • 3 BENCHMARKS
FakeAVCeleb is a novel Audio-Video Deepfake dataset that not only contains deepfake videos but respective synthesized cloned audios as well.
36 PAPERS • 2 BENCHMARKS
XD-Violence is a large-scale audio-visual dataset for violence detection in videos.
36 PAPERS • 1 BENCHMARK
AVSpeech is a large-scale audio-visual dataset comprising speech clips with no interfering background signals. The segments are of varying length, between 3 and 10 seconds long, and in each clip the only visible face in the video and audible sound in the soundtrack belong to a single speaking person. In total, the dataset contains roughly 4700 hours of video segments with approximately 150,000 distinct speakers, spanning a wide variety of people, languages and face poses.
35 PAPERS • NO BENCHMARKS YET
There exist previous works [6, 10] that constructed referring segmentation datasets for videos. Gavrilyuk et al. [6] extended the A2D [33] and J-HMDB [9] datasets with natural sentences; the datasets focus on describing the ‘actors’ and ‘actions’ appearing in videos, therefore the instance annotations are limited to only a few object categories corresponding to the dominant ‘actors’ performing a salient ‘action’. Khoreva et al. [10] built a dataset based on DAVIS [25], but the scales are barely sufficient to learn an end-to-end model from scratch
34 PAPERS • 3 BENCHMARKS
WildDeepfake is a dataset for real-world deepfakes detection which consists of 7,314 face sequences extracted from 707 deepfake videos that are collected completely from the internet. WildDeepfake is a small dataset that can be used, in addition to existing datasets, to develop more effective detectors against real-world deepfakes.
34 PAPERS • NO BENCHMARKS YET
JTA is a dataset for people tracking in urban scenarios by exploiting a photorealistic videogame. It is up to now the vastest dataset (about 500.000 frames, almost 10 million body poses) of human body parts for people tracking in urban scenarios.
32 PAPERS • 1 BENCHMARK
The BirdSong dataset consists of audio recordings of bird songs at the H. J. Andrews (HJA) Experimental Forest, using unattended microphones. The goal of the dataset is to provide data to automatically identify the species of bird responsible for each utterance in these recordings. The dataset contains 548 10-seconds audio recordings.
30 PAPERS • NO BENCHMARKS YET
The EgoGesture dataset contains 2,081 RGB-D videos, 24,161 gesture samples and 2,953,224 frames from 50 distinct subjects.
30 PAPERS • 2 BENCHMARKS
The EgoHands dataset contains 48 Google Glass videos of complex, first-person interactions between two people. The main intention of this dataset is to enable better, data-driven approaches to understanding hands in first-person computer vision. The dataset offers
OmniObject3D is a large vocabulary 3D object dataset with massive high-quality real-scanned 3D objects. OmniObject3D has several appealing properties:
URMP (University of Rochester Multi-Modal Musical Performance) is a dataset for facilitating audio-visual analysis of musical performances. The dataset comprises 44 simple multi-instrument musical pieces assembled from coordinated but separately recorded performances of individual tracks. For each piece the dataset provided the musical score in MIDI format, the high-quality individual instrument audio recordings and the videos of the assembled pieces.
EgoSchema is very long-form video question-answering dataset, and benchmark to evaluate long video understanding capabilities of modern vision and language systems. Derived from Ego4D, EgoSchema consists of over 5000 human curated multiple choice question answer pairs, spanning over 250 hours of real video data, covering a very broad range of natural human activity and behavior.
29 PAPERS • 2 BENCHMARKS
DeformingThings4D is a synthetic dataset containing 1,972 animation sequences spanning 31 categories of humanoids and animals. It provides 200 animations for humanoids and 1772 animations for animals.
28 PAPERS • 1 BENCHMARK
Dynamic FAUST extends the FAUST dataset to dynamic 4D data. It consists of high-resolution 4D scans of human subjects in motion, captured at 60 fps.
Activity recognition research has shifted focus from distinguishing full-body motion patterns to recognizing complex interactions of multiple entities. Manipulative gestures – characterized by interactions between hands, tools, and manipulable objects – frequently occur in food preparation, manufacturing, and assembly tasks, and have a variety of applications including situational support, automated supervision, and skill assessment. With the aim to stimulate research on recognizing manipulative gestures we introduce the 50 Salads dataset. It captures 25 people preparing 2 mixed salads each and contains over 4h of annotated accelerometer and RGB-D video data. Including detailed annotations, multiple sensor types, and two sequences per participant, the 50 Salads dataset may be used for research in areas such as activity recognition, activity spotting, sequence analysis, progress tracking, sensor fusion, transfer learning, and user-adaptation.
27 PAPERS • 1 BENCHMARK
The Gaming 3D Dataset (G3D) focuses on real-time action recognition in a gaming scenario. It contains 10 subjects performing 20 gaming actions: “punch right”, “punch left”, “kick right”, “kick left”, “defend”, “golf swing”, “tennis swing forehand”, “tennis swing backhand”, “tennis serve”, “throw bowling ball”, “aim and fire gun”, “walk”, “run”, “jump”, “climb”, “crouch”, “steer a car”, “wave”, “flap” and “clap”.
27 PAPERS • 2 BENCHMARKS
OpenLane is the first real-world and the largest scaled 3D lane dataset to date. The dataset collects valuable contents from public perception dataset Waymo Open Dataset and provides lane&closest-in-path object(CIPO) annotation for 1000 segments. In short, OpenLane owns 200K frames and over 880K carefully annotated lanes. The OpenLane Dataset is publicly released to aid the research community in making advancements in 3D perception and autonomous driving technology.
The Query-based Video Highlights (QVHighlights) dataset is a dataset for detecting customized moments and highlights from videos given natural language (NL). It consists of over 10,000 YouTube videos, covering a wide range of topics, from everyday activities and travel in lifestyle vlog videos to social and political activities in news videos. Each video in the dataset is annotated with: (1) a human-written free-form NL query, (2) relevant moments in the video w.r.t. the query, and (3) five-point scale saliency scores for all query-relevant clips.
27 PAPERS • 4 BENCHMARKS
A realistic dataset composed of 27 episodes from 6 popular TV series. The dataset spans over 16 hours of footage annotated with 30 action classes, totaling 6,231 action instances.
The Wide Multi Channel Presentation Attack (WMCA) database consists of 1941 short video recordings of both bonafide and presentation attacks from 72 different identities. The data is recorded from several channels including color, depth, infra-red, and thermal.
The Multi-Object and Segmentation (MOTS) benchmark [2] consists of 21 training sequences and 29 test sequences. It is based on the KITTI Tracking Evaluation 2012 and extends the annotations to the Multi-Object and Segmentation (MOTS) task. To this end, we added dense pixel-wise segmentation labels for every object. We evaluate submitted results using the metrics HOTA, CLEAR MOT, and MT/PT/ML. We rank methods by HOTA [1]. Our development kit and GitHub evaluation code provide details about the data format as well as utility functions for reading and writing the label files. (adapted for the segmentation case). Evaluation is performed using the code from the TrackEval repository.
26 PAPERS • 1 BENCHMARK
MOT20 is a dataset for multiple object tracking. The dataset contains 8 challenging video sequences (4 train, 4 test) in unconstrained environments, from crowded places such as train stations, town squares and a sports stadium. Image Source: https://motchallenge.net/vis/MOT20-04
26 PAPERS • 2 BENCHMARKS
MS-ASL is a real-life large-scale sign language data set comprising over 25,000 annotated videos.
Our dataset was made of videos from MSU Video Upscalers Benchmark Dataset, MSU Video Super-Resolution Benchmark Dataset and MSU Super-Resolution for Video Compression Benchmark Dataset. Dataset consists of real videos (were filmed with 2 cameras), video games footages, movies, cartoons, dynamic ads.
The MannequinChallenge Dataset (MQC) provides in-the-wild videos of people in static poses while a hand-held camera pans around the scene. The dataset consists of three splits for training, validation and testing.
26 PAPERS • NO BENCHMARKS YET
VALUE is a Video-And-Language Understanding Evaluation benchmark to test models that are generalizable to diverse tasks, domains, and datasets. It is an assemblage of 11 VidL (video-and-language) datasets over 3 popular tasks: (i) text-to-video retrieval; (ii) video question answering; and (iii) video captioning. VALUE benchmark aims to cover a broad range of video genres, video lengths, data volumes, and task difficulty levels. Rather than focusing on single-channel videos with visual information only, VALUE promotes models that leverage information from both video frames and their associated subtitles, as well as models that share knowledge across multiple tasks.
This is a dataset for a video super-resolution task. The dataset contains the most complex content for the restoration task: faces, text, QR-codes, car numbers, unpatterned textures, small details. Videos include different types of motion and different types of degradation: bicubic interpolation (BI) and Gaussian blurring and downsampling (BD). The resolution of all input video sequences is 480x320. Source: https://videoprocessing.ai/benchmarks/video-super-resolution.html Image Source: https://videoprocessing.ai/benchmarks/video-super-resolution.html
25 PAPERS • 1 BENCHMARK
A new multitask action quality assessment (AQA) dataset, the largest to date, comprising of more than 1600 diving samples; contains detailed annotations for fine-grained action recognition, commentary generation, and estimating the AQA score. Videos from multiple angles provided wherever available.
25 PAPERS • 2 BENCHMARKS
Contains 68,536 activity instances in 68.8 hours of first and third-person video, making it one of the largest and most diverse egocentric datasets available. Charades-Ego furthermore shares activity classes, scripts, and methodology with the Charades dataset, that consist of additional 82.3 hours of third-person video with 66,500 activity instances.
24 PAPERS • 1 BENCHMARK
Cityscapes-VPS is a video extension of the Cityscapes validation split. It provides 2500-frame panoptic labels that temporally extend the 500 Cityscapes image-panoptic labels. There are total 3000-frame panoptic labels which correspond to 5, 10, 15, 20, 25, and 30th frames of each 500 videos, where all instance ids are associated over time. It not only supports video panoptic segmentation (VPS) task, but also provides super-set annotations for video semantic segmentation (VSS) and video instance segmentation (VIS) tasks.
A large scale dataset with daily-living activities performed in a natural manner.
24 PAPERS • 2 BENCHMARKS