Program 2018

as of May 8, 2018

Instructions for Presentations

Talks: All contributed talks are scheduled for 15 min talk + 5 min discussion. Please adhere to the schedule.

We encourage using the presentation PC available in the lecture room (Windows with Acrobat viewer; Powerpoint is also available). Please report to the session chair/supporting staff 10 min before your session to upload your talk to the presentation PC.

Using your own laptop computer is possible but it is in your responsibility to ensure before the start of your session that your configuration works with the video projection system in the lecture hall.

Posters: Please bring a printed poster in A0 portrait format. Posters can be on display during the entire conference.

Tuesday, May 15, 2018

09:00–09:20 Opening

09:20–10:00 Contributed Session 1

  • Michael Opitz, Georg Waltner, Horst Possegger, Horst Bischof: Image Retrieval with BIER: Boosting Independent Embeddings Robustly

  • Christian Payer, Darko Stern, Horst Bischof, Martin Urschler: Multi-label Whole Heart Segmentation using Anatomical Label Configurations and CNNs

10:00–10:30 Coffee Break

10:30–11:50 Contributed Session 2

  • Ernst Schwartz, Karl-Heinz Nenning, Gregor Kasprian, Anna-Lisa Schuller, Lisa Bartha-Doering, Georg Langs: Multivariate Manifold Modelling of Functional Connectivity in Developing Language Networks

  • Roxane Licandro, Johannes Hofmanninger, Marc-André Weber, Bjoern Menze, Georg Langs: Early Predictors of Bone Infiltration in Multiple Myeloma Patients from T2 weighted MRI Images

  • Franz Thaler, Christian Payer, Darko Stern: Volumetric reconstruction from DRR images using CNN

  • Johannes Hofmanninger, Markus Krenn, Markus Holzer, Thomas Schlegl, Helmut Prosch, Georg Langs: Unsupervised Identification of Clinically Relevant Clusters in Routine Imaging Data

12:00–13:00 Lunch

13:30–14:30 Invited Talk

  • Marleen de Bruijne
    (Erasmus MC Rotterdam, The Netherlands / University of Copenhagen, Denmark):
    Machine learning imaging biomarkers

14:30–15:40 Coffee Break and Poster Session

  • Thomas Neff, Christian Payer, Darko Stern, Martin Urschler: Generative Adversarial Networks to Synthetically Augment Data for Deep Learning based Image Segmentation

  • Bernhard Blaschitz, Svorad Štolc, Doris Antensteiner: Multi-camera Array Calibration for Light Field Depth Estimation

  • Georg Wimmer, Michael Häfner, Andreas Uhl: CNN training using additionally training data extracted from frames of endoscopic videos

  • Jean-Baptiste Weibel, Timothy Patten, Michael Zillich: Bridging the gap between classical Robot Vision and Deep Learning

  • Simon Schreiberhuber, Johann Prankl, Markus Vincze: Towards ScalableFusion: Feasibility Analysis of a Mesh Based 3D Reconstruction

  • Thomas Lang: Page Segmentation and Region Classification Based on Region Bounding Boxes

  • Michael Breuß, Ashkan Mansouri Yarahmadi, Douglas Cunningham: The Convex-Concave Ambiguity in Perspective Shape from Shading

15:40–17:00 Contributed Session 3

  • Michael Breuß, Robert Dachsel, Laurent Hoeltgen: A Study of Spectral Expansion for Shape Correspondence

  • Martin Bähr, Michael Breuß, Robert Dachsel: Fast Solvers for Solving Shape Matching by Time Integration

  • Franz Schwanninger, Martin Welk: Image texture classification with morphological amoeba descriptors

  • Mario Prieshuber, Thomas Hütter, Stefan Katzenbeisser, Andreas Uhl: Depreciating Motivation and Empirical Security Analysis of Chaos-based Image and Video Encryption

17:15–18:15 General Assembly of AAPR/OAGM

19:00–22:00 Conference dinner (bus transfer from UMIT at 18:30, return to Kurhaus Hall approx. 22:15)

Wednesday, May 16, 2018

09:00–10:00 Invited Talk

  • Bjoern Menze
    (Technical University of Munich):
    Extracting and modeling information from medical images

10:00–10:30 Coffee Break

10:30–11:50 Contributed Session 4

  • Philipp Moll, Mathias Lux, Sebastian Theuermann, Hermann Hellwagner: A Network Traffic and Player Movement Model to Improve Networking for Competitive Online Games

  • Daniel Soukup, Thomas Pinetz: Reliably Decoding Autoencoders’ Latent Spaces for One-Class Learning Image Inspection Scenarios

  • Simon Brenner, Sebastian Zambanini, Robert Sablatnig: Detection of bomb craters in WWII aerial images

  • Manuel Keglevic, Robert Sablatnig: Semi-Automatic Retrieval of Toolmark Images

12:00–13:00 Lunch

13:20–14:40 Contributed Session 5

  • Christoph Heindl, Thomas Pönitz, Andreas Pichler, Josef Scharinger: Large Area 3D Human Pose Detection Via Stereo Reconstruction in Panoramic Cameras

  • Matthias Schörghuber, Marco Wallner, Roland Jung, Martin Humenberger, Margrit Gelautz: Vision-based Autonomous Feeding Robot

  • Christian Kapeller, Braulio Sespede, Matej Nezveda, Matthias Labschütz, Simon Flöry, Florian Seitner, Margrit Gelautz: A workflow for 3D model reconstruction from multi-view depth acquisitions of dynamic scenes

  • Rafael Weilharter, Fabian Schenk, Friedrich Fraundorfer: Globally Consistent Dense Real-Time 3D Reconstruction from RGBD Data

14:45–15:15 Coffee Break

15:20–16:00 Contributed Session 6

  • Alexander Grabner, Peter M. Roth, Vincent Lepetit: Efficient 3D Pose Estimation and 3D Model Retrieval

  • Georg Poier, David Schinagl, Horst Bischof: Being lazy at labelling for pose estimation

16:00–16:20 Best Paper Award and Closing