New AI computer vision software released by Fujitsu to ease behavior recognition training
Fujitsu has developed a new artificial intelligence-based method of performing behavioral analysis on video footage, which it says can recognize a range of subtle and complex human activities without large amounts of training data, according to a company announcement.
The new “Actlyzer” technology was developed by Fujitsu Laboratories and the Fujitsu Research and Development Center, and combines about 100 basic actions it is pretrained for modularly to identify more complex behaviors, such as acting suspiciously or considering a purchase. In contrast, Fujitsu says, deep learning technologies typically rely on huge amounts of video to train the recognition of individual behaviors, which means they take several months to be introduced in the field.
The systems accuracy for recognizing the 100 basic actions is 90 percent or higher on average, according to Fujitsu.
Suggested potential uses for the technology include automatic detection of suspicious activity, product interest surveys based on recognized purchase behavior, and training applications by comparing the skills of workers with different levels of experience in factories.
Actlyzer will be offered in Japan during calendar 2019, and is expected to be commercialized internationally as part of the Fujitsu Human Centric AI Zinrai suite of products.
To test out the technology, Fujitsu conducted an experiment, and was able to identify eight kinds of suspicious behavior in 21 different kinds of indoor and outdoor video with only a day to create the rules by combining basic actions.
Fujitsu has been carrying out research into several different areas of computer vision, including the development of an AI technology for recognizing subtle changes in facial expression.