Generative AI & WebXR Engineer

Favourite Quotes:

"In contrast to the predominant vision architecture, convolutional neural networks (CNNs), the transformer architecture contains no built-in inductive prior on the locality of interactions and is therefore free to learn complex relationships among its inputs. However, this generality also implies that it has to learn all relationships, whereas CNNs have been designed to exploit prior knowledge about strong local correlations within images."
-Taming Transformers for High-Resolution Image Synthesis

"For this experiment, we consider a dataset consisting of two shapes, bob, a genus-1 rubber duck, and spot, a genus-0 cow."
-MeshSDF

"For many tasks, like object transfiguration, the desired output is not even well-defined. We assume there is some underlying relationship between the domains - for example, that they are two different renderings of the same underlying scene - and seek to learn that relationship."
-Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

"Compared to a generative model for 2D images, a 3D generative model naturally disentangles pose and appearance."
-Convolutional Generation of Textured 3D Meshes

"My broad view is that each time another "magic" tech comes along, e.g. macro assemblers, high-level compilers, high-level compilers with proof-based whole-system optimisation, ... etc ... the next human contribution just moves up the abstraction level a bit."
-Some Guy, Hacker News, on the topic of DALL-E 2

"One day, in retrospect, the years of struggle will strike you as the most beautiful."
-Sigmund Freud