Date: 02/06/2021

Panasonic develops AI dataset of human activity in homes

Panasonic develops two advanced artificial intelligence dataset related technologies which are accepted to CVPR2021 (IEEE Conference on Computer Vision and Pattern Recognition).

Panasonic developed a new dataset called "Home Action Genome" collecting human's daily activities in their homes using various types of sensors including cameras, microphones and thermal sensors. Panasonic says it has constructed and released the world's largest multimodal dataset for living spaces. This dataset which is comparatively big can be used as training data for machine learning and AI research to support people in living space.

Panasonic has also developed a cooperative learning technology for hierarchical activity recognition in multimodal and multiple viewpoints. Panasonic says by applying this technology, we can learn consistent features between different viewpoints, sensors, hierarchical behaviors, and detailed behavior labels, and thus improve the recognition performance of complex activities in living spaces.
This technology is developed through a research conducted in collaboration between the Digital AI Technology Center, Technology Division, and the Stanford Vision and Learning Lab at Stanford University.

The other AI technology Panasonic developed is a new machine learning technology that automatically performs optimal data augmentation according to the distribution of training data.
"There are many cases in our main business areas, where it is difficult to apply AI technology because of the limitations of the available data. By applying this technology, the tuning process of data augmentation parameters can be eliminated, and the parameters can be adjusted automatically." says Panasonic in its release.