Let me join you! Real-time F-formation recognition by a socially aware robot

TL;DR

This paper introduces a real-time, socially aware robot system that detects social groups and predicts approach angles using skeletal key points, CRF, and SVM, outperforming existing methods significantly.

Contribution

The novel architecture combines skeletal key point estimation with CRF and SVM models for accurate, real-time social group detection and approach angle prediction in robotic applications.

Findings

Achieved 91% accuracy in group and outlier detection.

Outperformed state-of-the-art by 29% in formation detection.

Outperformed state-of-the-art by 55% in combined formation and approach angle detection.

Abstract

This paper presents a novel architecture to detect social groups in real-time from a continuous image stream of an ego-vision camera. F-formation defines social orientations in space where two or more person tends to communicate in a social place. Thus, essentially, we detect F-formations in social gatherings such as meetings, discussions, etc. and predict the robot's approach angle if it wants to join the social group. Additionally, we also detect outliers, i.e., the persons who are not part of the group under consideration. Our proposed pipeline consists of -- a) a skeletal key points estimator (a total of 17) for the detected human in the scene, b) a learning model (using a feature vector based on the skeletal points) using CRF to detect groups of people and outlier person in a scene, and c) a separate learning model using a multi-class Support Vector Machine (SVM) to predict the exact F-formation of the group of people in the current scene and the angle of approach for the viewing robot. The system is evaluated using two data-sets. The results show that the group and outlier detection in a scene using our method establishes an accuracy of 91%. We have made rigorous comparisons of our systems with a state-of-the-art F-formation detection system and found that it outperforms the state-of-the-art by 29% for formation detection and 55% for combined detection of the formation and approach angle.