FIDAC: An Easy-to-use Pipeline to Extract and Interpret Interpersonal Distance From Video

2026-07-27

AI Summary

This study addresses the limitations of existing video analysis methods in accurately extracting and interpreting interpersonal distances, which stem from a lack of reliable and user-friendly tools. To overcome this challenge, the authors propose and open-source FIDAC, a novel system that integrates the strengths of multiple face detection models with face trajectory tracking, a human-in-the-loop correction pipeline, and a depth-aware distortion calibration mechanism to significantly enhance spatial position estimation accuracy. By doing so, FIDAC enables high-precision, interpretable analysis of proxemic space and substantially improves the accuracy and practicality of extracting interpersonal distance data from ordinary video footage. This advancement provides robust support for research in human behavior and social interaction.

Abstract

The distance between persons reveals significant information about their perception of each other. However, such information is not easily extractable and interpretable from video input. We developed an open-sourced library, Facial Interpersonal Distance Analysis and Coding (FIDAC) that transforms facial detection results into actionable data about location and interpersonal distance. This tool merges data from multiple open-source facial detection models, strategically compensating for gaps in any individual model. In addition, we include methods for more accurate tracking, such as a pipeline for human coding of the selection of faces and a benchmarking tool to reduce depth distortion. For next steps, we plan on building upon FIDAC by evaluating its effectiveness at measuring interpersonal distance at various depths and orientations while further integrating features of proxemic analysis such as synchrony into its software.

Problem

• interpersonal distance
• video analysis
• proxemics
• facial detection
• distance interpretation

Innovation

• interpersonal distance
• facial detection fusion
• depth distortion correction
• proxemic analysis
• video-based behavioral coding