Action Recognition In Videos

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 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 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

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

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/

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

The Sports-1M dataset consists of over a million videos from YouTube. The videos in the dataset can be obtained through the YouTube URL specified by the authors. Approximately 7% (as of 2016) of the videos have been removed by the YouTube uploaders since the dataset was compiled. However, there are still over a million videos in the dataset with 487 sports-related categories with 1,000 to 3,000 videos per category. The videos are automatically labelled with 487 sports classes using the YouTube Topics API by analyzing the text metadata associated with the videos (e.g. tags, descriptions). Approximately 5% of the videos are annotated with more than one class. Source: Review of Action Recognition and Detection Methods Image Source: Computer Vision for Sports

This paper introduces the pipeline to scale the largest dataset in egocentric vision EPIC-KITCHENS. The effort culminates in EPIC-KITCHENS-100, a collection of 100 hours, 20M frames, 90K actions in 700 variable-length videos, capturing long-term unscripted activities in 45 environments, using head-mounted cameras. Compared to its previous version (EPIC-KITCHENS-55), EPIC-KITCHENS-100 has been annotated using a novel pipeline that allows denser (54% more actions per minute) and more complete annotations of fine-grained actions (+128% more action segments). This collection also enables evaluating the "test of time" - i.e. whether models trained on data collected in 2018 can generalise to new footage collected under the same hypotheses albeit "two years on". The dataset is aligned with 6 challenges: action recognition (full and weak supervision), action detection, action anticipation, cross-modal retrieval (from captions), as well as unsupervised domain adaptation for action recognition. For each challenge, we define the task, provide baselines and evaluation metrics.

The Kinetics-600 is a large-scale action recognition dataset which consists of around 480K videos from 600 action categories. The 480K videos are divided into 390K, 30K, 60K for training, validation and test sets, respectively. Each video in the dataset is a 10-second clip of action moment annotated from raw YouTube video. It is an extensions of the Kinetics-400 dataset. Source: Learning to Localize Actions from Moments Image Source: https://towardsdatascience.com/downloading-the-kinetics-dataset-for-human-action-recognition-in-deep-learning-500c3d50f776

The 20BN-SOMETHING-SOMETHING 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 108,499 videos, with 86,017 in the training set, 11,522 in the validation set and 10,960 in the test set. There are 174 labels. ⚠️ Attention: This is the outdated V1 of the dataset. V2 is available here. Source: https://20bn.com/datasets/something-something/v1 Image Source: https://20bn.com/datasets/something-something/v1

AVA is a project that provides audiovisual annotations of video for improving our understanding of human activity. Each of the video clips has been exhaustively annotated by human annotators, and together they represent a rich variety of scenes, recording conditions, and expressions of human activity. There are annotations for: - Kinetics (AVA-Kinetics) - a crossover between AVA and Kinetics. In order to provide localized action labels on a wider variety of visual scenes, authors provide AVA action labels on videos from Kinetics-700, nearly doubling the number of total annotations, and increasing the number of unique videos by over 500x. - Actions (AvA Actions) - the AVA dataset densely annotates 80 atomic visual actions in 430 15-minute movie clips, where actions are localized in space and time, resulting in 1.62M action labels with multiple labels per human occurring frequently. - Spoken Activity (AVA ActiveSpeaker, AVA Speech). AVA ActiveSpeaker: associates speaking activity with a visible face, on the AVA v1.0 videos, resulting in 3.65 million frames labeled across ~39K face tracks. AVA Speech densely annotates audio-based speech activity in AVA v1.0 videos, and explicitly labels 3 background noise conditions, resulting in ~46K labeled segments spanning 45 hours of data. Image Source: https://www.researchgate.net/profile/Paolo_Napoletano/publication/309327222/figure/fig1/AS:419620126248965@1477056642346/Sample-images-from-the-Aesthetic-Visual-Analysis-AVA-database-sorted-by-their-aesthetic.png

Extended GTEA Gaze+ EGTEA Gaze+ is a large-scale dataset for FPV actions and gaze. It subsumes GTEA Gaze+ and comes with HD videos (1280x960), audios, gaze tracking data, frame-level action annotations, and pixel-level hand masks at sampled frames. Specifically, EGTEA Gaze+ contains 28 hours (de-identified) of cooking activities from 86 unique sessions of 32 subjects. These videos come with audios and gaze tracking (30Hz). We have further provided human annotations of actions (human-object interactions) and hand masks. The action annotations include 10325 instances of fine-grained actions, such as "Cut bell pepper" or "Pour condiment (from) condiment container into salad". The hand annotations consist of 15,176 hand masks from 13,847 frames from the videos. Source: http://cbs.ic.gatech.edu/fpv/ Image Source: http://cbs.ic.gatech.edu/fpv/

The PKU-MMD dataset is a large skeleton-based action detection dataset. It contains 1076 long untrimmed video sequences performed by 66 subjects in three camera views. 51 action categories are annotated, resulting almost 20,000 action instances and 5.4 million frames in total. Similar to NTU RGB+D, there are also two recommended evaluate protocols, i.e. cross-subject and cross-view. Source: Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation Image Source: https://www.icst.pku.edu.cn/struct/Projects/PKUMMD.html

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. Image Source: https://assembly-101.github.io/

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. Source: What and How Well You Performed? A Multitask Learning Approach to Action Quality Assessment

Ego4D is a massive-scale egocentric video dataset and benchmark suite. It offers 3,025 hours of daily life activity video spanning hundreds of scenarios (household, outdoor, workplace, leisure, etc.) captured by 855 unique camera wearers from 74 worldwide locations and 9 different countries. The approach to collection is designed to uphold rigorous privacy and ethics standards with consenting participants and robust de-identification procedures where relevant. Ego4D dramatically expands the volume of diverse egocentric video footage publicly available to the research community. Portions of the video are accompanied by audio, 3D meshes of the environment, eye gaze, stereo, and/or synchronized videos from multiple egocentric cameras at the same event. Furthermore, a host of new benchmark challenges are presented, centered around understanding the first-person visual experience in the past (querying an episodic memory), present (analyzing hand-object manipulation, audio-visual conversation, and social interactions), and future (forecasting activities). By publicly sharing this massive annotated dataset and benchmark suite, the aim is to push the frontier of first-person perception. Description from: Facebook AI Paper: Ego4D: Around the World in 3,000 Hours of Egocentric Video GitHub: https://github.com/EGO4D

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. Source: Online Action Detection

Jester Gesture Recognition dataset includes 148,092 labeled video clips of humans performing basic, pre-defined hand gestures in front of a laptop camera or webcam. It is designed for training machine learning models to recognize human hand gestures like sliding two fingers down, swiping left or right and drumming fingers. Source: Gesture Recognition Dataset: Jester

EgoExoLearn is a fascinating dataset designed to bridge the gap between egocentric and exocentric views of procedural activities. 1. What Is EgoExoLearn? EgoExoLearn is a large-scale dataset that emulates how humans learn by observing others. It focuses on the process of asynchronous demonstration following. Participants in the dataset record egocentric videos as they perform tasks. These videos are guided by exocentric-view demonstration videos. In simpler terms, imagine someone watching a demonstration video (from an external perspective) and then replicating the same task while recording their own point-of-view video. 2. Dataset Details: EgoExoLearn dataset spans 120 hours and covers scenarios from daily life and specialized laboratories. It contains: - Egocentric videos: These are recorded by individuals executing tasks. - Demonstration videos: These show the same tasks from an external viewpoint. - Gaze data: High-quality gaze information accompanies the videos. - Multimodal annotations: Detailed annotations provide context and insights. 3. Applications and Benchmarks: The EgoExoLearn dataset serves as a playground for modeling the human ability and thus provides a playground to bridge asynchronous procedural actions from different viewpoints. It enables new benchmarks such as: - Cross-view association: Linking actions observed from different perspectives. - Cross-view action planning: Anticipating and planning actions based on both ego and exo views. - Cross-view referenced skill assessment: Evaluating skills across viewpoints. Using Exo-view demonstrations as guidance for better ego-view skill assessment. 4. Why Is It Important? - Understanding how we map others' activities into our own point of view is a fundamental human skill. - EgoExoLearn paves the way for creating AI agents capable of seamlessly learning by observing humans in the real world. (1) EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric .... https://arxiv.org/html/2403.16182v1. (2) EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric .... https://egoexolearn.github.io/. (3) EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric .... https://arxiv.org/abs/2403.16182. (4) EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric .... https://allainews.com/item/egoexolearn-a-dataset-for-bridging-asynchronous-ego-and-exo-centric-view-of-procedural-activities-in-real-world-2024-03-26/. (5) undefined. https://github.com/OpenGVLab/EgoExoLearn/.

Covers 5 generic driving scenarios, with a total of 25 distinct action classes. It contains more than 15K full HD, 5s long videos acquired in various driving conditions, weathers, daytimes and environments, complemented with a common and realistic set of sensor measurements. This amounts to more than 2.25M frames, each annotated with an action label, corresponding to 600 samples per action class. Source: VIENA2: A Driving Anticipation Dataset

Operating rooms (ORs) are complex, high-stakes environments requiring precise understanding of interactions among medical staff, tools, and equipment for enhancing surgical assistance, situational awareness, and patient safety. Current datasets fall short in scale, realism and do not capture the multimodal nature of OR scenes, limiting progress in OR modeling. To this end, we introduce MM-OR, a realistic and large-scale multimodal spatiotemporal OR dataset, and the first dataset to enable multimodal scene graph generation. MM-OR captures comprehensive OR scenes containing RGB-D data, detail views, audio, speech transcripts, robotic logs, and tracking data and is annotated with panoptic segmentations, semantic scene graphs, and downstream task labels. Further, we propose MM2SG, the first multimodal large vision-language model for scene graph generation, and through extensive experiments, demonstrate its ability to effectively leverage multimodal inputs. Together, MM-OR and MM2SG establish a new benchmark for holistic OR understanding, and open the path towards multimodal scene analysis in complex, high-stakes environments. Paper: https://arxiv.org/abs/2503.02579

Is one of the largest egocentric datasets in the object search task with eyetracking information available Source: Deep Future Gaze: Gaze Anticipation on Egocentric Videos Using Adversarial Networks

Existing image/video datasets for cattle behavior recognition are mostly small, lack well-defined labels, or are collected in unrealistic controlled environments. This limits the utility of machine learning (ML) models learned from them. Therefore, we introduce a new dataset, called Cattle Visual Behaviors (CVB), that consists of 502 video clips, each fifteen seconds long, captured in natural lighting conditions, and annotated with eleven visually perceptible behaviors of grazing cattle. By creating and sharing CVB, our aim is to develop improved models capable of recognizing all important cattle behaviors accurately and to assist other researchers and practitioners in developing and evaluating new ML models for cattle behavior classification using video data. The dataset is presented in the form of following three sub-directories. 1. raw_frames: contains 450 frames in each sub folder representing a 15 second video taken at a frame rate of 30 FPS. 2. annotations: contains the json files corresponding to the raw_frames folder. There is one json file for each video, that contains the bounding-box annotations for each cattle in the video and its associated behavior, and 3. CVB_in_AVA_format: contains the CVB data in the AVA dataset format.

We present a new simulated dataset for pedestrian action anticipation collected using the CARLA simulator. To generate this dataset, we place a camera sensor on the ego-vehicle in the Carla environment and set the parameters to those of the camera used to record the PIE dataset (i.e., 1920x1080, 110° FOV). Then, we compute bounding boxes for each pedestrian interacting with the ego vehicle as seen through the camera's field of view. We generated the data in two urban environments available in the CARLA simulator: Town02 and Town03. The total number of simulated pedestrians is nearly 55k, equivalent to 14M bounding boxes samples. The critical point for each pedestrian is their first point of crossing the street (in case they will eventually cross) or the last bounding box coordinates of their path in the opposite case. The crossing behavior represents 25% of the total pedestrians. We balanced the training split of the dataset to obtain labeled sequences crossing/non-crossing in equal parts. We used sequence-flipping to augment the minority class (i.e., crossing behavior in our case) and then undersampled the rest of the dataset. The result is a total of nearly 50k pedestrian sequences. Next, the pedestrian trajectory sequences were transformed into observation sequences of equal length (i.e., 0.5 seconds) with a 60% overlap for the training splits. The TTE length is between 30 and 60 frames. It resulted in a total of nearly 220k observation sequences.

Daily Activity Recordings for Artificial Intelligence (DARai, pronounced "Dahr-ree") is a multimodal, hierarchically annotated dataset constructed to understand human activities in real-world settings. DARai consists of continuous scripted and unscripted recordings of 50 participants in 10 different environments, totaling over 200 hours of data from 20 sensors including multiple camera views, depth and radar sensors, wearable inertial measurement units (IMUs), electromyography (EMG), insole pressure sensors, biomonitor sensors, and gaze tracker. To capture the complexity in human activities, DARai is annotated at three levels of hierarchy: (i) high-level activities (L1) that are independent tasks, (ii) lower-level actions (L2) that are patterns shared between activities, and (iii) fine-grained procedures (L3) that detail the exact execution steps for actions. The dataset annotations and recordings are designed so that 22.7% of L2 actions are shared between L1 activities and 14.2% of L3 procedures are shared between L2 actions. The overlap and unscripted nature of DARai allows counterfactual activities in the dataset. Experiments with various machine learning models showcase the value of DARai in uncovering important challenges in human-centered applications. Specifically, we conduct unimodal and multimodal sensor fusion experiments for recognition, temporal localization, and future action anticipation across all hierarchical annotation levels. To highlight the limitations of individual sensors, we also conduct domain-variant experiments that are enabled by DARai's multi-sensor and counterfactual activity design setup. The code, documentation, and dataset are available at the dedicated DARai website.

First of its kind paired win-fail action understanding dataset with samples from the following domains: “General Stunts,” “Internet Wins-Fails,” “Trick Shots,” & “Party Games.” The task is to identify successful and failed attempts at various activities. Unlike existing action recognition datasets, intra-class variation is high making the task challenging, yet feasible.