Hi all,
We will have two student presentations upcoming Wednesday (07.02.2024) at 1:00 pm.
First, Julia Töws will be presenting her final Bachelor thesis talk.
Then, Marie Schmidt will be presenting her first Bachelor thesis talk.
You can find the thesis titles and abstracts below.
The presentation will take place in E1.7, room 001.
Best regards,
Marie
————————————Julia Töws——————————————
Title:
Co-creative 3D Modeling in VR: Investigating AI Representation Modes
Abstract:
Recent HCI research has highlighted the domain of Extended Reality (XR) as a superior interface for 3D modeling compared to traditional 2D interfaces. Concurrently, research for generative AI is rapidly advancing, with increased development specifically in the field of 3D object generation. This work aims to shine light on a little-researched intersection of the aforementioned fields: Co-creative systems for 3D modeling in XR. Co-creative systems - software allowing human users to work together with AI to collaboratively work on shared artifacts - are often evaluated in terms of functionality, with little focus on users' perceptions of and subjective experiences with the system. In this work, we examine whether different aspects of visualizing a co-creative AI and its contributions can affect such perceptual factors. We conducted an empirical study investigating the three aspects of AI contribution highlighting, incremental contribution visualization, and AI embodiment, using a simplified implementation of a co-creative 3D modeling system in VR. Quantitative and qualitative data was collected from questionnaires, system logs, observations, and interviews. Our analysis shows differing effects across the three visualization aspects: Highlighting was not found to affect any of our measures on participants' perceptions and behavior. Incremental visualization was primarily perceived as bothersome, though evaluations tended to differ when embodiment was additionally enabled. Embodiment showed the strongest effects on human-AI relationships and attribution of contributions to the AI. Though the generalizability of our findings is limited, this work raises attention to a number of HCI-related questions for XR and co-creativity research to explore in the future.
————————————Marie Schmidt————————————
Title:
Exploring Fabric Actuation and Deformation Through Magnetic Force
Abstract:
Clothing is more than just a functionality-based item, it is also a form of self-expression. Fabric actuation can offer exciting ways to support this need for self-expression through movement and shape-change. Current fabric actuation techniques often use Shape-Memory-Alloys, since their wire-like structure affords easy integration as well as versatile actuation movement. However, their actuation is rather slow-paced, limiting the possibility for rapid or high frequency actuation. While pneumatic or robotic approaches offer fast-paced actuation, the required hardware is pretty bulky, making it difficult to integrate into fabric. To address this, my bachelor thesis is exploring the use of magnetic force as an actuation mechanism for fabrics. Through a systematic investigation, I identify and evaluate different actuation possibilities with magnetic force, culminating in a design workshop to explore potential applications