Based on the Vision AI-based analysis technology developed independently by P2ACH AI,
it realizes efficiency of retail media, operation, and marketing through analysis in space.
The AI Agent developed by P2ACH AI goes beyond simple visualization of offline behavioral data collected through Vision AI, embodying an Agentic Intelligence capable of autonomous understanding, reasoning, and decision-making.
This technology structurally interprets data context by integrating real-time vision analytics with historical data, KPI benchmarks, and operational rules, all grounded in a proprietary ontology that defines offline environments across people, space, media, products, and behavioral dimensions.
By combining an LLM-based natural language reasoning engine with rule-based analytical logic, the AI Agent interprets user intent within the dashboard and automatically retrieves and analyzes the required data. It performs root-cause analysis, identifies key drivers of change, and detects anomalies.
As a result, the system moves beyond delivering raw metrics to providing concrete action recommendations that directly influence advertising operations, spatial layout, and content strategies in physical offline environments—enabling intelligent, decision-oriented support rather than passive reporting
For Ultra-Precise Gaze Analysis
Extrinsic Calibration, which we have implemented, is one of the key technologies used for autonomous driving.
This advanced mathematical technology accurately estimates and reconstructs the 3D position of objects by correcting distortions and errors within the camera's field of view (FOV). By precisely calculating the location, orientation, and coordinates of target objects, it even enables sophisticated gaze analysis to determine exactly where a person is looking.
High Performance
'Detection' identifies targets like people or vehicles through AI algorithms, while 'Tracking' continuously follows those detected objects in real time.
Specifically, Detection recognizes the emergence and presence of target objects. And if Tracking performance is unstable, it results in Frame Missing, which can cause the system to misidentify the same object as a different one upon re-detection, lowering analysis reliability. Our AI algorithms achieve high-performance tracking to minimize Frame Missing, ensuring maximum data integrity for ad performance, foot traffic, and heatmap analysis.
Validation of Technical Excellence through Accredited Testing Institutions.
We have confirmed technical excellence by significantly exceeding performance standards across all categories, verifying the precision of our AI models in ad-viewing area detection and achieving high accuracy in facial recognition factors such as gender and age estimation.
Delivering Efficiency and Cost-effectiveness
This technology allows a single Edge AI Box to connect two cameras and run analysis models simultaneously.
By integrating two cameras with a single AI Box, it significantly enhances the scalability of the analysis area. Furthermore, it enables simultaneous analysis of both the front and back areas of double-sided devices, providing the functionality of two units for the cost of a single AI Box.
Retail Media & Insight
Person Re-Identification (Re-ID) Re-ID is a sophisticated AI technology that maintains a consistent identity for individuals across multiple camera views, ensuring precise tracking and behavioral analysis.
Re-identification: Examples of Multi-camera Re-ID & Staff Recognition.
Universal High-Performance Edge AI
Versatile Multi-OS Support and High-Performance Edge AI Solutions
Our advertising performance analysis service uses only anonymized data and does not process personal information. The data anonymization process is as follows: