Causal inference strategies in human vision (04)

• Subject Area: General, Cognitive and Mathematical Psychology ; Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing ; Cognitive, Systems and Behavioural Neurobiology
• Term: from 2017 to 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 276693517

Much of the success of artificial vision systems is typically achieved in highly controlled settings, when the data generating distribution does not change between training and testing. Human vision, on the other hand, is robust to out of distribution changes in sensor noise, illumination, view point, occlusion, or different environments. In this project, we explore the hypothesis that the visual system combines causal inference with generative modelling strategies in order to produce robust perception. In an extension to the first proposal, we hereby focus on how temporal gestalt principles facilitate an object-based scene understanding in humans and machines.

• DFG Programme: Collaborative Research Centres
• Subproject of: SFB 1233: Robust Vision - Inference Principles and Neural Mechanisms
• Applicant Institution: Eberhard Karls Universität Tübingen
• Project Heads: Professor Dr. Matthias Bethge; Professor Dr. Bernhard Schölkopf; Professor Dr. Felix A. Wichmann