Visual Bank Launches Qlean Dataset Featuring Ski Mogul Video and Image Collections

Visual Bank's New Offering: Ski Mogul Data Set

Visual Bank Inc., based in Minato-ku, Tokyo, has expanded its AI learning data solutions with the introduction of the 'Qlean Dataset', recently adding a ski mogul video and image dataset. This move is part of their broader strategy to enhance their portfolio of AI development datasets through their subsidiary, Amano Images.

The 'Qlean Dataset' serves researchers and companies alike by offering original data suitable for commercial use. It features a flexible combination of ready-to-use data materials based on specific needs such as precision and deadlines. The datasets also provide options for both annotated and non-annotated data, with capacity for customization based on individual requirements.

Visual Bank has partnered with organizations like the Chiba Lotte Marines and Toyo Keizai Inc., as well as leveraging both domestic and international networks to further enrich their dataset offerings. These initiatives will significantly reduce the burden of data collection and maintenance in the AI development field, thus accelerating project timelines and outputs.

Overview of the Ski Mogul Video and Image Dataset

The newly launched ski mogul dataset includes video and image footage capturing male and female athletes performing mogul skiing. Here are some key details about the dataset:

- Subjects: Two male Japanese athletes in their 20s
- Environment: Outdoor at a ski resort
- Capture Angles: Simultaneously captured from two angles

Video Dataset

- Data Size: 2.83 GB
- Number of Files: 26
- Format: MP4
- Total Duration: 14 minutes and 23 seconds
- Sample URL: Video Dataset Sample

Image Dataset

- Data Size: 24.31 GB
- Number of Files: 6,373
- Format: JPEG
- Sample URL: Image Dataset Sample

Potential Applications of the Ski Mogul Dataset

1. Sports Motion Analysis and Impact Absorption Research: The dataset captures the unique movements involved in mogul skiing, including jumps, turns, and landings from multiple angles. It's ideal for analyzing impact absorption mechanics and estimating joint movements in fields like sports science and medical research.

2. Airborne Position Estimation and Form Optimization: The footage captures intricate changes in posture during jumps and tricks, making it a valuable resource for training airborne position estimation models and form improvement Ai.

3. Behavior Recognition and Sliding Motion Analysis: The dataset includes continuous actions such as sliding, turning, and slowing down, ideal for the development of action recognition or motion tracking Ai, particularly in snowy environments where visibility can vary.

4. Judging Criteria and Trick Evaluation AI Development: The dataset comprehensively covers the three key judging criteria in mogul skiing: speed, turns, and air (aerial position), making it suitable for developing AI-supported judging models for competitions.

5. VR/AR Skiing Experience and Video Production Ai Development: The multi-angle data can facilitate realistic skiing experiences within VR/AR environments, suitable for developing educational content, entertainment, and competition simulation.

Distinguishing Features of Qlean Dataset

- Research and Commercial Use: The datasets come with releases and consent from all subjects involved, complying with international privacy policies, ensuring the data's usability in both research and commercial projects.
- AI Data Recipes: Employing a unique provision framework allows for speedy data acquisition while maximizing return on investment, thus minimizing initial investment costs for users.
- Custom Dataset Creation: For datasets not featured in the 'AI Data Recipes', Visual Bank offers custom-built datasets tailored to specific requirements, ensuring unique data needs are met effectively.

Qlean Dataset Academia Support Program

Visual Bank is also launching a free data provision program aimed at universities, research institutions, and non-profit tech development teams. They offer access to over 500,000 high-quality and rights-cleared learning data points across more than 80 categories, including images, audio, video, and text. This initiative aims to address the challenge many researchers face regarding access to quality learning data.

For more details on the Qlean Dataset and partnership opportunities, visit:

Visual Bank, under CEO Masayuki Nagai, is dedicated to creating and providing a next-gen data infrastructure that maximizes AI development capabilities. Their mission is to 'unleash the potential of all data', with significant contributions through various AI-driven products and projects. Their commitment has also led to participation in Japan's national research and development program, GENIAC, enabling further acceleration of their initiatives.