Fetching the paper…

Reading the bibliography…

2019

Real-time Facial Surface Geometry from Monocular Video on Mobile GPUs

Kartynnik, Yury, Ablavatski, Artsiom, Grishchenko, Ivan et al.

Original
PDF
Post on X

Understand

We present an end-to-end neural network-based model for inferring an approximate 3D mesh representation of a human face from single camera input for AR applications.

The relatively dense mesh model of 468 vertices is well-suited for face-based AR effects.
The proposed model demonstrates super-realtime inference speed on mobile GPUs (100-1000+ FPS, depending on the device and model variant) and a high prediction quality that is comparable to the variance in manual annotations of the same image.

Real-time Facial Surface Geometry from Monocular Video on Mobile GPUs · Around

Built on

Recursively generated B-spline surfaces on arbitrary topological meshes
Edwin Catmull and Jim Clark · 1978
Earlier work this paper cites.
A morphable model for the synthesis of 3D faces
Volker Blanz and Thomas Vetter · 1999
Earlier work this paper cites.
What is the space of camera response functions?
Michael D. Grossberg and Shree K. Nayar · 2003
Earlier work this paper cites.

Similar

1 € filter: A simple speed-based low-pass filter for noisy input in interactive systems
Géry Casiez, Nicolas Roussel, and Daniel Vogel · 2012
Cited alongside, same era.
How far are we from solving the 2D & 3D face alignment problem? (and a dataset of 230,000 3D facial landmarks)
Adrian Bulat and Georgios Tzimiropoulos · 2017
Cited alongside, same era.
Morphable face models - an open framework
Original
Thomas Gerig · 2017
Cited alongside, same era.

Then

https://github.com/tensorflow/tensorflow/tree/master/tensorflow/lite/delegates/gpu
TFLite on GPU · 2019
Closest in time.
Blazeface: Sub-millisecond neural face detection on mobile GPUs
Authors · 2019
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…