Rebecca Mobbs

I am a PhD researcher in Artificial Intelligence at the School of Computer Science and Mathematics, Kingston University London. My research focuses on the application of multimodal artificial intelligence for human-centred computing, with particular interests in emotion analysis, facial behaviour understanding, and DeepFake detection.

My doctoral research explores how artificial intelligence can understand, generate, and reason about human emotions and behaviour. My initial PhD project investigated emotion-controlled speech-driven facial animation, exploring AI's ability to modify emotions in both facial expressions and speech within video content. This work examined how emotional information can be transferred across visual and auditory modalities to create more expressive and controllable virtual humans.

Building on this foundation, my current research investigates the use of large vision-language models for zero-shot DeepFake detection. I explore how emotional and facial dynamics can be used to identify manipulated media without task-specific training. Through this work, I aim to develop more explainable and generalisable approaches to digital media forensics.

Prior to my current research, I completed a Msc in psychology, which sparked my interest in the intersection of human behaviour and artificial intelligence. This interdisciplinary background informs my research approach, combining insights from psychology, affective computing, computer vision, and machine learning.

My broader research interests include generative AI, multimodal learning, affective computing, digital media authenticity, human-AI interaction, and emotionally intelligent AI systems. I am particularly interested in developing AI technologies that can better understand, model, and reason about human emotions and behaviour.