- Media recognition
- Media data analysis
Object detection and tracking technology in video
This platform technology offers a good balance between processing speed and accuracy, and can be used in a wide range of applications.
- Detects and tracks specified objects such as people or vehicles in video, and senses in the motion of those objects.
- Achieves efficient processing because two independent technologies are combined into a single structure.
- Even in videos with low frame rates, which have difficulty tracking objects, objects can be tracked accurately using appearances of objects, because the process is based on the same structure as object detection.
Applications
- Security and wide area monitoring
- Marketing
- Approvals
- Visual presentation of work processes
Various other sensing functions that use videos
Benchmarks, strengths, and track record
- Used in Toshiba Digital Solutions Corporation facial recognition AI “Kaometa.” (in Japanese)
- Used in Toshiba Lighting & Technology Corporation’s ViewLED Solution (tlt.co.jp). (in Japanese)
Inquiries
Please include the title “Toshiba AI Technology Catalog: Technology for detecting and tracking regular objects in video” or the URL in the inquiry text.
Please note that because this technology is currently the subject of R&D activities, immediate responses to inquiries may not be possible.
References:
- Kobayashi, Shibata; “Simultaneous estimate of multiple object tracking and object detection using Spatiotemporal Attention, MIRU2020.
- Automatic recognition of athletes with 98.1% accuracy, using image analysis AI (in Japanese)
- The Toshiba Co-creation Center “Creative Circuit” Implements Video Recognition AI to Promote Value Creation with Co-creation Partners -Accelerating the social implementation of AI technology by sharing images using Toshiba’s latest video recognition AI- | Corporate Research & Development Center | Toshiba (global.toshiba)