NEXT WEEK: NEW APPROACHES TO 3D VISION, 1st-4th November 2021, hosted by the Royal Society

This Royal Society online meeting brings together researchers from computer vision, animal vision, and human vision to explore recent developments in 3D vision. Speakers reflect both academia and industry, with representatives from DeepMind, Facebook Reality Labs, Google, and Microsoft Research.

Date/Time: 1st-4th November 2021, 3-6pm (3-6:30pm on the 4th), UK Time Please note that the USA/UK time difference is 1hr less than usual

Website: https://royalsociety.org/science-events-and-lectures/2021/11/3d-vision/

Registration (Free): https://www.eventbrite.co.uk/e/new-approaches-to-3d-vision-registration-1657...

DAY ONE (1st Nov) - Seeing Beyond SLAM Chair: Andrew Fitzgibbon (Microsoft)

Session One: Neural Scene Representation SM Ali Eslami (DeepMind): “Neural priors, neural encoders and neural renderers” Ida Momennejad (Microsoft Research): “Multi-scale predictive representations and human-like RL”

Session Two: Perception-Action Loop Sergey Levine (UC Berkeley and Google): “Generalization in data-driven control” Andrew Glennerster (University of Reading): “Understanding 3D vision as a policy network”

DAY TWO (2nd Nov) - Animals in Action Chair: Matteo Carandini (University College London)

Session One: Locating Prey and Rewards Jenny Read (Newcastle University): “Stupid stereoscopic algorithms that still work” Aman Saleem (University College London): “Visual processing in the brain during navigation”

Session Two: Navigation in 3D Space Kate Jeffery (University College London): “The cognitive map of 3D space: not as metric as we thought?” Gily Ginosar (Weizmann Institute of Science): “Locally ordered representation of 3D space in the entorhinal cortex”

DAY THREE (3rd Nov) - Experiencing Space Chair: Mar Gonzalez-Franco (Microsoft Research)

Session One: Theories of Visual Space Dhanraj Vishwanath (University of St Andrews): “Tripartite encoding of visual 3D space” Paul Linton (City, University of London): “New approaches to visual scale and visual shape”

Session Two: Challenges for Virtual Reality Sarah Creem-Regehr (University of Utah): “Perception and Action in Virtual and Augmented Reality” Douglas Lanman (Facebook Reality Labs): “Engineering challenges for realistic displays”

DAY FOUR (4th Nov) - Grasping the World Chair: Jody Culham (Western University)

Session One: One Visual Stream or Two? Fulvio Domini (Brown University): “A novel non-probabilistic model of 3D cue integration explains both perception and action” Irene Sperandio (University of Trento): “Dissociations between perception and action in size-distance scaling”

Session Two: 3D Space and Visual Impairment Ione Fine (University of Washington): “Do you hear what I see? How do early blind individuals experience object motion?” Ewa Niechwiej-Szwedo (University of Waterloo): “The role of binocular vision in the development of visuomotor control and performance of fine motor skills”

Session Three: Panel Discussion Andrew Fitzgibbon (Microsoft), Matteo Carandini (University College London), Mar Gonzalez-Franco (Microsoft Research), and Jody Culham (Western University), share their thoughts on the future of 3D vision with an interactive Question and Answer session.

Organisers: Michael Morgan FRS (City, University of London), Paul Linton (City, University of London), Jenny Read (Newcastle University), Dhanraj Vishwanath (University of St Andrews), Sarah Creem-Regehr (University of Utah), Fulvio Domini (Brown University)