Responding to feedback from local governments and businesses, the eagerly awaited video version is now available – PIXTA begins selling 'Bear Video Dataset' for machine learning
PIXTA launches 'Bear Video Dataset' for machine learning.
📋 Article Processing Timeline
- 📰 Published: March 28, 2026 at 00:25
- 🔍 Collected: March 28, 2026 at 21:59 (21h 33m after Published)
- 🤖 AI Analyzed: April 15, 2026 at 02:19 (412h 19m after Collected)
PIXTA (Pixsta Co., Ltd., Shibuya-ku, Tokyo, Representative Director and President: Daisuke Komata, Tokyo Stock Exchange Standard: 3416), a marketplace for photo, illustration, video, and music materials, is pleased to announce the launch of its new 'Bear Video Dataset' as a 'Machine Learning Image/Video Data Provision Service.' This follows the highly successful 'Bear Image Dataset' released last year. In response to recent serious bear damage and fulfilling strong demands from development sites to 'learn and verify walking patterns and movements using videos, not just still images,' we have developed a video dataset. This will support the development of more advanced AI for motion detection and behavior analysis.
[Apology and Correction] There was an error in the displayed sales price in the press release distributed today at 11:00 AM. We sincerely apologize and make the following correction:
(Incorrect) 99,000 yen (tax included)
(Correct) 198,000 yen (tax included)
◆ PIXTA Machine Learning Image/Video Data Provision Service: https://pixta.jp/machinelearning-dataset
■ Bear Video Dataset Overview
|
Name |
Bear Video Dataset |
|
Number of Data |
50 video clips... |