1

Neural Lens Modeling

We propose a neural lens model that is versatile, easy to use and can be trivially used in gradient-based optimization pipelines for point projection and rendering. We also propose a new dataset for evaluation of lens and calibration models and a new strategy to create marker boards for calibration.

NeuWigs: A Neural Dynamic Model for Volumetric Hair Capture and Animation

We create a drivable and accurate hair representation using neural rendering primitives and temporal models.

Human View Synthesis using a Single Sparse RGB-D Input

We create a novel view synthesis system for humans using only a single RGB-D sensor.

TAVA: Template-free Animatable Volumetric Actors

We create rigged and skinned animation-ready, editable actor representations from a 3D skeleton and multi-view data.

Efficient Estimation of Boundary Integrals for Path-Space Differentiable Rendering

Better and faster estimation of boundary integrals improves physics-based differentiable rendering.

HVH: Learning a Hybrid Neural Volumetric Representation for Dynamic Hair Performance Capture

We create a metrically accurate hair representation using neural rendering primitives and optical flow.

Virtual Elastic Objects

We create virtual objects that visually and physically resemble their real counterparts.

Neural 3D Video Synthesis

We propose a novel approach for 3D video synthesis that is able to represent multi-view video recordings of a dynamic real-world scene in a compact, yet expressive representation that enables high-quality view synthesis and motion interpolation.

Self-supervised Neural Articulated Shape and Appearance Models

We propose a novel approach for learning a representation of the geometry, appearance, and motion of a class of articulated objects given only a set of color images as input.

Non-rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene from Monocular Video

We propose a joined training for neural radiance fields and a deformation field which enables us to do 4D reconstruction over time, even using only a single camera.