Institute of Computer Graphics and Vision

Vision · Learning · Optimization

Research Highlights

	Belief Propagation Reloaded: Learning BP-Layers for Labeling Problems
	Total Deep Variation for Linear Inverse Problems
	Learned Collaborative Stereo Refinement
	Self-Supervised Learning for Stereo Reconstruction on Aerial Images
	End-to-end training of hybrid CNN-CRF models for stereo
	A first-order primal-dual algorithm for convex problems with applications to imaging

Projects/Funding

ERC Starting Grant

HOMOVIS
High-level Prior Models for Computer Vision

BioTechMed Graz Flagship Project

ILearnHeart
Image-based Learning in Predictive Personalized Models of Total Heart Function

BioTechMed Graz Flagship Project

MIDAS
Inflammatory Mechanisms in Diabetes Uncovered by Tissue Imaging and Machine Learning

NIH

LOVIR
Learning an Optimized Variational Network for Medical Image Reconstruction

FFG Bridge

RE:Color
Efficient coloring of movies in cinema quality based on novel methods in machine learning

DES Lab

Dependable Embedded Systems

FFG IKT der Zukunft

Instant Avatar
Instant Photorealistic Generation of Human Avatars with Mobile Devices

Digital Material Valley Styria

MikroMet-AI

© Lunghammer - TU Graz

Team Menu