Menu
Sign In Search Podcasts Charts People & Topics Add Podcast API Pricing
Podcast Image

UCL Minds

Lunch Hour Lecture | Faces of the Future: AI's Journey Beyond the realm of strangeness

26 Feb 2024

Description

About the Lecture: The extent to which a face appears alive or lifeless has long been a topic in psychology, with the idea that more humanlike-looking faces achieve greater familiarity until a point is reached at which subtle imperfections give a sensation of strangeness – the uncanny valley effect. The uncanny valley effect term describes the sense of discomfort or unease we experience when we encounter a robot with certain human-like characteristics. With rapid advances in technology, AI-generated faces are now widely available and are being used for both helpful and criminal purposes, from finding missing children to transmitting political misinformation via fake social media accounts. In this talk, Dr Krumhuber will give a brief historical perspective on how we have overcome the uncanny valley with AI faces that are now indistinguishable from human faces. Also, Dr Krumhuber will present her recent work which found that White AI faces are judged as human more often than actual human facesβ€”a phenomenon we term AI hyperrealism. About the Speaker Eva Krumhuber is associate professor in the Department of Experimental Psychology at University College London. Much of her work is concerned with the empirical investigation of the socio-cognitive and affective processes in human perception and behaviour. This includes research on facial expressions, especially morphological and dynamic features and their role in emotion interpretation. More recently, she started exploring commonalities and differences in human and machine classification of emotions, with a particular focus on how various elicitation methods (i.e., posed, spontaneous, naturalistic) influence recognition accuracy. She has published widely within the field of psychology and computer science.

Audio
Featured in this Episode

No persons identified in this episode.

Transcription

This episode hasn't been transcribed yet

Help us prioritize this episode for transcription by upvoting it.

0 upvotes
πŸ—³οΈ Sign in to Upvote

Popular episodes get transcribed faster

Comments

There are no comments yet.

Please log in to write the first comment.