---
title: "Gesture Recognition Dataset"
description: "10,000+ videos 5 hand gestures"
url: "https://unidata.pro/datasets/gesture-recognition/"
date_modified: "2025-12-09T10:54:29+03:00"
language: "ru-RU"
---
This dataset offers videos of clearly demonstrated hand gestures for hand gesture recognition, movement recognition, and hand gesture detection, providing labeled training data that supports recognition systems and hand tracking for building AI models that accurately identify hand movements, finger movements, and specific gestures to improve user interfaces, object detection, and sign recognition applications.

## Структура датасета

### Секция с числами

**Numbers list:**

- **Number:** 10,000+ — **Text:** videos
- **Number:** 5 — **Text:** hand gestures

### Секция тултипов

**Tooltip items:**

- **Name:** Gesture Recognition
- **Name:** Detection
- **Name:** Human-Computer Interaction
- **Name:** Computer Vision
- **Name:** Machine learning

### Dataset Info

**Таблица с данными:**

| Characteristic | Data |
| --- | --- |
| Description | Each video shows a person clearly demonstrating a single hand gesture |
| Data types | Video |
| Tasks | Detection, Classification, Recognition |
| Number of video | 10,000+ |
| Labeling | Metadata(id, gender) |
| Gender | Male, female |
| Type of hand gesture | One, four, small, fist, and me. |

**Слайдер с медиа:**

- **Видео в сладйер:** <https://unidata.pro/wp-content/uploads/2024/12/one.webm>
- **Видео в сладйер:** <https://unidata.pro/wp-content/uploads/2024/12/four.webm>

**Ссылка на сэмпл:** [Download sample](https://drive.google.com/drive/folders/1UzkqmcLKsU6ylQ74kjfi3S5EEM9nSS8r)

### Technical  characteristics

**Таблица с данными:**

| Characteristic | Data |
| --- | --- |
| File extension | MP4 |

**Source and collection methodology:** Source and collection methodology. Data was collected by UniData team by using the crowdsourcing service

### Dataset Use Cases - слайдер

**Карточки индустрий:**

- **Индустрия:** Human–Computer Interaction — **Заголовок:** Gesture-Based Control Systems — **Текст:** Gesture Recognition Dataset helps develop touchless interfaces by training machine learning models to interpret hand gestures, finger movements, and body motions. It supports gesture detection and hand tracking for applications like AR/VR environments, smart homes, and automotive systems, enabling more natural and responsive user interaction experiences.
- **Индустрия:** Healthcare & Rehabilitation — **Заголовок:** Monitoring and Movement Analysis — **Текст:** This dataset can be used to build recognition systems for tracking patient progress during physical therapy or rehabilitation exercises. Analyzing dynamic hand and arm movements helps medical software identify proper gesture patterns, ensuring accurate monitoring and assisting in personalized recovery programs.
- **Индустрия:** Technology & AI Development — **Заголовок:** Training Gesture Recognition Models — **Текст:** Developers use the dataset as high-quality training data for neural networks and deep learning models that recognize hand poses, signs, and movements. It enhances computer vision systems in gesture identification and object detection, improving accuracy in sign language recognition, wearable devices, and motion-based interfaces.
- **Индустрия:** Gaming & Entertainment — **Заголовок:** Gesture-Controlled Gameplay and Immersive Experiences — **Текст:** The dataset supports creating gesture-based gaming environments and virtual reality experiences. It allows developers to train models that accurately interpret gestures and hand motions, improving player engagement through natural, intuitive controls and seamless integration with 3D models and motion tracking technologies.

### Фак

**Заголовок FAQs:** FAQs

**Перечень вопросов:**

- **Вопрос:** What types of annotations are provided? — **Ответ:** The dataset includes metadata annotations detailing participant ID and gender, which are useful for analyzing gesture performance across demographics. These labels help refine learning algorithms and improve model generalization for gesture recognition tasks.
- **Вопрос:** Is it possible to request a custom dataset? — **Ответ:** Yes. Unidata supports the creation of custom gesture datasets based on specific research needs, device setups, or gesture categories. You can request tailored training datasets that align with unique requirements for gesture recognition and hand motion detection models.
- **Вопрос:** Can I request a sample of the dataset before purchasing? — **Ответ:** Yes. Unidata provides free sample data so you can test the hand gesture recognition quality, labeling format, and compatibility with your machine learning pipeline before purchasing the full dataset.
- **Вопрос:** Do Unidata datasets follow GDPR or other data privacy regulations? — **Ответ:** Yes. All Unidata datasets fully comply with GDPR and other relevant data protection laws. Each dataset is collected from legally permissible sources, ensuring privacy and ethical handling of all recorded participants and content.
- **Вопрос:** How are Unidata datasets stored? — **Ответ:** All Unidata datasets are securely stored on AWS cloud infrastructure, ensuring both high availability and strong data protection. The storage system complies with ISO 27001 and ISO 27701 standards, providing a reliable environment for AI training datasets.
- **Вопрос:** How long does it take to receive the dataset? — **Ответ:** After submitting your purchase request, Unidata will confirm your details and issue the necessary documentation. Once the agreement and payment are finalized, the dataset is securely delivered within 3–10 business days.
- **Вопрос:** Why is a single gesture recorded in each video? — **Ответ:** Each video focuses on one clearly demonstrated gesture, allowing AI models to learn distinct motion patterns without interference from unrelated actions. This improves classification accuracy and simplifies supervised model training.
- **Вопрос:** Can this dataset be used for robotics and physical AI? — **Ответ:** Yes. Robots can use gesture recognition models trained on this dataset to understand human commands and improve human-robot interaction. It is particularly valuable for service robots, collaborative robots, and intelligent automation systems.