Eric-Tuan Le

I am currently in my final year doing a PhD in the Smart Geometry Processing Group, which is part of the School of Computer Science at University College London. Under the co-supervision of Iasonas Kokkinos and Niloy J. Mitra, my research revolves around Deep Learning, Computer Vision, and Computer Graphics. Specifically, I have been focusing on 3D representation and reconstruction.

I hold two Master's degrees: one in Applied Mathematics from CentraleSupelec, where I graduated with the highest honors and achieved the best overall grade in my class, and another in Machine Learning and Computer Vision (Master MVA) from Ecole Normale Supérieure Paris-Saclay (formerly known as ENS Cachan).

In 2020, I had the opportunity to intern remotely at Adobe Research in San Jose, where I worked on primitives' fitting for large 3D point clouds. Then, in 2022, I interned at Snap Inc. London, where my project involved 3D body mesh reconstruction from 2D images with improved 2D reprojection accuracy. Currently, I am interning at Meta AI in London until December 2023.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github

We introduce StyleMorph, a 3D-aware generative model that disentangles 3D shape, camera pose, object appearance, and background appearance for high quality image synthesis. We chain 3D morphable modelling with deferred neural rendering by performing an implicit surface rendering of “Template Object Coordinates” (TOCS).

We introduce Softmesh, a fully differentiable pipeline to transform a 3D point cloud into a probabilistic mesh representation that allows us to directly render 2D images.

We present Cascaded Primitive Fitting Networks (CPFN) that relies on an adaptive patch sampling network to assemble detection results of global and local primitive detection networks. As a key enabler, we present a merging formulation that dynamically aggregates the primitives across global and local scales.

We train deeper and more accurate point processing networks by introducing three modular point processing blocks that improve memory consumption and accuracy. By combining these blocks, we design wider and deeper point-based architectures.

Source code from Jon Barron's website