#Japan #Tokyo #Qlean Dataset #Visual Bank #AI Data Recipe

Visual Bank Introduces American Football Dataset

Visual Bank Inc., based in Minato, Tokyo, under the leadership of CEO Masayuki Nagai, is expanding its offerings through a subsidiary, Amana Images. The company has launched the American Football video and image dataset as part of their AI learning data solution called the Qlean Dataset. This dataset aims to facilitate research and commercial AI development by providing high-quality, flexible data resources, further enriching their unique lineup known as AI Data Recipe.

The Qlean Dataset and AI Data Recipe

The AI Data Recipe represents a collection of original data available for commercial use within the Qlean Dataset. It stands out for its ability to provide flexible combinations of readily available data materials based on various application needs, accuracy, and deadlines. It also accommodates both annotated and non-annotated data, allowing for personalized changes and expansions to fit specific requirements.

Collaborations with partners like the Chiba Lotte Marines and Toyo Keizai Inc., along with an expanding network both domestically and internationally, enhance the dataset offerings. The aim is to significantly reduce the burden of data collection and organization in AI development, thus accelerating the overall development process.

Details on the American Football Video and Image Dataset

Common Details of the Video and Image Dataset

• - Contents: This dataset includes a variety of actions related to American football captured through video and image data.
• - Subjects: 10 Japanese male football players in their twenties.
• - Shooting Environment: Conducted outdoors on an American football field.
• - Shooting Angles: Simultaneous filming from two angles.
• - Actions Captured: Various movements including running, play-action, option plays, and special plays.

Video Dataset

• - Data Size: 4.14 GB
• - Number of Data Pieces: 28
• - Data Format: mp4
• - Filming Duration: 14 minutes and 33 seconds

Sample Details

Image Dataset

• - Data Size: 43.29 GB
• - Number of Data Pieces: 6,623
• - Data Format: jpeg

Sample Details

Use Case Scenarios for the American Football Dataset

• - Team Tactics Analysis and Behavioral Recognition AI Development

This dataset is ideal for developing AI related to teamwork, allowing analysis of multi-player formations and actions from various angles, making it suitable for academia's sports strategy analysis or sports tech companies developing video analysis services.

• - Motion Estimation AI Research

With high-resolution captures of contact-based actions like running and tackling, it's useful for analyzing different positions' forms or researching motion to prevent injuries. This can aid research in sports science and exercise engineering, optimizing bodily movements and fatigue estimation AI development.

• - Play Pattern Recognition and Video Analysis AI Development

Covering a variety of tactics within a single team, this dataset is advantageous for AI that recognizes play types through video analysis or for creating analytical models for match situations. It can also be used in developing tactical support applications and game analysis tools.

• - Competitive Scoring and Performance Evaluation Support

By allowing comparisons of movements based on positions and roles, it supports AI systems for evaluating players and training assistance, making it well-suited for coaching and automated tagging of competitive data.

• - VR/AR and Sports Education Training Reproduction

The multi-angle video data can be used for reproducing matches in VR/AR environments or for educational simulations, also facilitating the development of interactive AI learning materials to train decision making and play selection skills.

Features of Qlean Dataset

The datasets provided by Qlean Dataset are tailored for both research and commercial utilization. Consent for data usage has been obtained from all subjects, ensuring compliance with privacy policies across various countries, allowing users to engage in research and commercial activities with confidence.

The unique AI Data Recipe format enables rapid data procurement, minimizing initial investments. Additionally, for customized requests, datasets not available within the AI Data Recipe can be built to meet specific requirements, leveraging the capabilities of Qlean Dataset.

For inquiries, please visit the Qlean Dataset contact form.

Academic Support Program

Visual Bank is also launching a free dataset offering program as part of its support for academia, targeted at universities, research institutions, and non-profit technical development teams. With over 80 varieties totaling more than 500,000 data points, this initiative addresses the challenge of acquiring high-quality, rights-cleared training data for research settings.

For more details, visit: Qlean Dataset Academia Support.

Visual Bank Inc. aims to build and provide next-generation data infrastructure to maximize AI development capabilities, advocating for unleashing the potential of all data. Besides Qlean Dataset, they offer tools such as THE PEN, an AI-assisted tool for manga artists.

The company is recognized under the national research and development program, GENIAC, accelerating its societal implementation efforts.

CEO: Masayuki Nagai
Address: 6F C-Cube Minamiaoyama Building 7-1-7 Minato, Tokyo 107-0062
Visual Bank Corporate Site
Amana Images Corporate Site