Google Research Fellowship for Aron Monszpart

UCL-CS Research Student Aron Monszpart has been awarded the Google PhD Fellowship in Machine Perception to continue exploring how computers can learn about their environment from observing how humans use it. Aron is a member of the Virtual Environments and Computer Graphics Research Group.

Aron's research topic is 3D scene understanding with unseen priors. We as humans use many other cues than just vision to interpret our environment, something that's still lacking in scene understanding performed by machines. Aron's research investigates how simple assumptions about the environment can yield great returns in terms of the quality of knowledge that can be inferred. For example, real scenes should obey laws of physics or be recognised as environments that were designed by or are used by humans. Using physics laws to disambiguate partial 3D information is particularly useful, since we have reached a clear understanding of the processes shaping our immediate environment during the last 400 years!

In the end, in any real-life situation, one doesn't only see surfaces and materials "floating" in space, but can also guess why they are there and what purpose they have. This allows one to efficiently interpret the environment, despite parts being invisible (e.g., occluded) or even impossible to see (e.g., physical forces). 

More about Aron's research can be found here.

Aron has just returned from Mountain View, California where he attended a "PhD fellowship summit".