Playtag launches mono_xyz to analyze behavior via natural language

[UP! START] A new technology has emerged that allows users to ask questions in natural language and receive quantitative behavioral analysis from mobile phone footage to existing CCTV, all without additional equipment or coding. The solution is Playtag's mono_xyz, a behavior foundation technology.

According to Playtag on the 21st, the company has launched mono_xyz (hereafter referred to as mono), an AI solution that performs precise analysis by answering natural language questions about actions within a video.

Mono is a conversational behavior analysis solution that, upon receiving mobile phone or CCTV footage as input, provides quantitative data on what a person in the video did and how they performed it.

Overcoming the limitations of existing video analysis solutions, which could only detect simple, pre-designated rules such as intrusions or falls, mono analyzes the continuous behavioral context and intentions of individuals in the video and presents them as data without any separate pre-settings.

In particular, because users can immediately check behavioral data by asking questions exactly as humans communicate—without additional equipment or complex coding processes—the technology is expected to drastically improve work efficiency in various fields handling video data.

Playtag has been developing a behavior foundation model using 7,000,000 hours of video behavior data. Mono was created to answer users' natural language questions using this unrivaled model. Mono accurately identifies even minute behavioral changes within complex visual environments.

Currently, mono is conducting pilot programs at five institutions and business sites. In senior care settings, it is being applied to detect falls and verify care routines; in early childhood physical education facilities, it is used to analyze children's educational engagement and activity levels to provide evidence for activity improvement and link to recommended activities; and in construction and manufacturing sites, it is being utilized for worker movement analysis and safety rule compliance verification.

Playtag plans to gradually expand the scope of mono's application to include customer movement analysis in retail stores, human-robot collaboration safety in smart factories, and monitoring developmental and cognitive changes in caregiving settings.

"Through Storyline, we have accumulated the world's largest scale of 3D behavior analysis data in actual living spaces, and mono is a model born from that data," said Jang Hyeon-su, CEO of Playtag. "Since the limitation of video analysis development has been the data rather than the algorithms, we will present a new standard for behavior analysis AI that allows people to ask and answer questions just as they communicate with video."