Region-Feature Combinations in Facial Analysis

Summary

Facial analysis has significant applications in human-computer interaction and multimedia communication research. In this paper, we aim to investigate the impact of facial regions and features in the facial analysis tasks. In particular, we compare the performances of various facial features extracted from different regions in two specific tasks, namely, age group prediction and twin recognition. Furthermore, we also evaluate the fusion of handcrafted features and deep learned features in both tasks. Through the experiments, we highlight different impacts of facial regions and facial features combination in facial analysis tasks.

ITCD Dataset

We collect a new dataset named Identical Twins Celebrity Dataset (ITCD), for evaluation purposes of the twin recognition.

Stats: 

- Popular twin celebrities from different countries, i.e., United States, China, England, Brazil, etc. - Images of the twins with a constraint that both twin siblings appear in the same image. 

- Images with a large amount of noise or occlusion are discarded. 

- 1060 face photos from 60 twins of celebrities.

Download link [.zip]