Lars GTC Europe talk on the Dynamics of Generative Adversarial Networks is now available.
We are looking for a PhD student working on interpretable representations.
For the first time, the prestigious IEEE PAMI Young Researcher Award goes to a German researcher.
Application Deadline: Nov 15, 2018
The International Max Planck Research School for Intelligent Systems is now hiring our 2019 generation of Ph.D. students. They will contribute to world-leading research in areas such as machine learning, computer vision, robot dynamics and control, micro- and nano-robotics, and haptics.
Eshed Ohn-Bar from CMU receives Humboldt Fellowship and will join AVG in December as a post-doctoral researcher!
Am 1. März begann Geiger die Professur für "Learning-based Computer Vision" an der über 500 Jahre alten, renommierten Universität Tübingen. Geiger leitet weiterhin eine Forschungsgruppe am MPI-IS in Tübingen.
Andreas GTC Europe talk on Deep Models for 3D Reconstruction is now available.
The German Pattern Recognition Award is awarded once a year to one young researcher in computer vision, pattern recognition or machine learning at an age of 35 years or less and sponsored by the Daimler AG with 5000€.
3D reconstruction from multiple 2D images is an inherently ill-posed problem. Prior knowledge is required to resolve ambiguities and probabilistic models are desirable to capture the ambiguities in the reconstructed model. In this talk, I will present two recent results tackling these two aspects. First, I will introduce a probabilistic framework for volumetric 3D reconstruction where the reconstruction problem is cast as inference in a Markov random field using ray potentials. Our main contribution is a discrete-continuous inference algorithm which computes marginal distributions of each voxel's occupancy and appearance. I will show that the proposed algorithm allows for Bayes optimal predictions with respect to a natural reconstruction loss. I will further demonstrate several extensions which integrate non-local CAD priors into the reconstruction process. In the second part of my talk, I will present a novel framework for deep learning with 3D data called OctNet which enables 3D CNNs on high-dimensional inputs. I will demonstrate the utility of the OctNet representation on several 3D tasks including classification, orientation estimation and point cloud labeling. Finally, I will present an extension of OctNet called OctNetFusion which jointly predicts the space partitioning function with the output representation, resulting in an end-to-end trainable model for volumetric depth map fusion.