Computational Modeling of Facial Images and Their Description

A workshop co-located with Nodalida 2019

September 30th, Turku, Finland

Faces 2019 will be held at the 22nd Nordic Conference on Computational Linguistics (NoDaLiDa 2019) Conference, Turku University, Turku, Finland, on 30 September 2019.

The workshop will provide a forum for presenting and discussing current research focusing on the way human faces are perceived, understood and described by humans, as well as the way computational models represent (and therefore ‘understand’) human faces and generate descriptions of faces, or facial images corresponding to descriptions, for different purposes and applications.

Recent research on multimodal analysis of images and text or the generation of image descriptions has focussed on general data sets which might contain faces as a subset. However processing human faces at the Vision-Language interface presents unique challenges. Descriptions of faces are frequent in human communication, for example when one seeks to identify an individual or distinguish one person from another. They are also pervasive in descriptive or narrative text. Depending on the context, they may focus on physical attributes, or incorporate inferred characteristics and emotional elements.

The ability to adequately model and describe faces is interesting for a variety of language technology applications, e.g., conversational agents and interactive narrative generation, as well as forensic applications in which faces need to be identified or generated from textual or spoken descriptions. Such systems need to process the images associated with human faces together with their linguistic descriptions, therefore the research needed to develop them is placed at the interface between vision and language research, a new cross-disciplinary area which has received considerable attention in recent years, e.g. through the series of workshops on Vision and Language organised by the European Network on Integrating Vision and Language (iV&L Net), the 2015–2018 Language and Vision Workshops, or the more recent Workshops on Shortcomings in Vision and Language.

Human faces are being studied by researchers from different research communities, including those working in computer vision, natural language understanding, natural language generation, as well as cognitive science, cognitive psychology, multimodal communication and embodied conversational agents. The workshop aims to reach out to all these communities to explore the many different aspects of research on human faces and foster cross-disciplinary synergy.