Selfies, posies, and group pictures

• Main Author: Ahmad Aleena
• Other Authors: Deepu Rajan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Social media analytics; Selfie detection; Resnet-50; Transfer learning; Deepface; RetinaFace; VGG16; Group emotion analysis; GUI; Valence-arousal
• Online Access: https://hdl.handle.net/10356/175329
• Institution: Nanyang Technological University

With the increasing popularity of social media apps, social media analytics have a wide range of applications in discerning user behaviour, preferences and interactions. My final year project titled "Selfies, posies, and group photos" delves into the intersection of deep learning, computer vision, and psychology to explore the relationship between an individual’s psychological traits in real life versus the portrayal on their Instagram. In collaboration with the Psychology Department of the School of Social Sciences (SSS) at Nanyang Technological University, the project aims to use deep learning to analyze an individual’s Instagram posts. Employing state-of-art Convolutional Neural Networks (CNN) of Resnet50 for selfie classification and RetinaFace, DeepFace and VGG16 , this project introduces a comprehensive methodology for individual and group picture analysis. An “Image Emotion Analyzer” GUI is also implemented using python programming language to enable a user to gain valuable insights from an image. This research not only enhances the capabilities of image analytics, but also bridges the gap between computer vision and psychology by laying a foundation for deeper understanding of an individual’s crafted personality on social media.