Applied Vision
Organizers
Harald Ganster*, Florian Kleber#, Roland Perko*, Gernot Stübl+, Martina Uray*
* Joanneum Research
# TU Wien
+ FH OOE Wels
Workshop Description
The Austrian Association for Pattern Recognition (OAGM) invites authors and guests to participate in the Applied Vision Workshop as part of the AIRoV event, held in person at the Technical University of Leoben. The workshop provides a platform for presentation and discussion of research progress as well as current projects within the OAGM community. We additionally focus on interdisciplinary research of computer vision and pattern analysis in the context of Applied Vision. The aim is to bring together Austrian and nearby groups, researchers, and students working on interdisciplinary topics for discussion and establishing potential collaborations.
General topics of interest include but are not limited to:
- 3D vision, stereo, and structure from X
- Animal and plant phenotyping
- Cultural heritage
- Cyber-physical systems
- Document analysis
- Embedded and mobile computer vision/machine learning
- Explainability, ethics, and fairness in computer vision/machine learning
- Image denoising, restoration, and enhancement
- Image, video, and multimodal retrieval
- Industrial assistance systems
- Inspection and quality assurance
- Motion and tracking
- Multi- and hyperspectral image analysis
- Object/scene detection, recognition, and categorization
- Pattern analysis (visual and non-visual data)
- Precision agriculture and precision forestry
- Remote sensing and earth observation
- Sensors and information fusion
- Video analysis and event recognition
- Vision datasets, benchmarking, and performance evaluation
Program
Invited Talk
Bridging Domains: Practical Computer Vision from Defence to Space
The threshold for high-performance computer vision and intelligent systems is dropping rapidly. AI-assisted development tools and the widespread proliferation of powerful, affordable processing hardware are enabling a broad range of applications that were until recently reserved for well-funded research projects. One domain where this shift is becoming obvious to everyone is the modern battlefield, where autonomous robots are being adopted at unprecedented scale (from UAV to UUV). Their capabilities grow with every iteration, yet truly autonomous operations remain an unsolved challenge and the bottleneck of real time communication remains. Space as another frontier is experiencing a similar disruption: Earth observation satellites are gaining increasingly powerful onboard processing capabilities, while in-orbit servicing missions and deep-space probes place ever greater demands on autonomous visual navigation and decision-making. Across both domains, the core technical challenges converge: real-time inference on edge hardware under strict size, weight, and power constraints, and reliable operation without human-in-the-loop. This talk presents how Fantana, an Austrian SME operating at the intersection of space, defence and industry, navigates this rapidly changing landscape and transfers lessons learned between domains.
Thomas Riel: is the founder and General Manager of Fantana GmbH in Vienna, where he bridges the gap between academic research and field-deployed systems in defence, space, and industrial settings. His current work spans electro-optical drone detection and tracking, onboard image processing for Earth observation satellites, and machine-learning-driven metrology for industrial production lines. He holds a PhD in Opto-Mechatronics from TU Wien and is actively involved in ESA-, EDA- and EDF-funded R&D initiatives, collaborating with Austrian and European aerospace partners.
Schedule
First Session: Wednesday April 15, 2026, 09:00–10:30
- Roland Perko (JOANNEUM RESEARCH), Stefanie Onsori-Wechtitsch (JOANNEUM RESEARCH), Helmut Neuschmied (JOANNEUM RESEARCH), Peter Schallauer (JOANNEUM RESEARCH), Katharina Hofer-Schmitz (JOANNEUM RESEARCH), Michaela Stolz (biohelp), "AI-Based Optimization of Roadside Mowing Operations in Austria" (Short Paper)
- Christian Schweighofer (FH OOE), "Obstacle Detection Pipeline using Monocular Depth Estimation in Mobile Robotics" (Short Paper)
- Alexander Dvorak (TU Wien), Michael Nowak (TU Wien), Tessa Pulli (TU Wien), Markus Vincze (TU Wien), "GraspGen+HSR: Adapting Simulation-Trained 6-DoF Grasping to Real Service Robots Without Retraining" (Full Paper)
- Matvey Ivanov (TU Wien), Markus Vincze (TU Wien), Peter Hönig (TU Wien), "Diffusion Edge Detection Of Texture-less Objects" (Full Paper)
- Thummanoon Kunanuntakij (TU Wien), Dominik Schörkhuber (TU Wien), Margrit Gelautz (TU Wien), "Synthetic Skeletal Pose Pre-training to Mitigate Data Scarcity in In-Cabin 2D-to-3D Pose Lifting" (Short Paper)
Second Session: Wednesday April 15, 2026, 11:00–12:30
- Ido Akov (Austrian Institute of Technology), Roman Pflugfelder (Austrian Institute of Technology), Daniel Cremers (Technical University of Munich), "Fourier contrast optimization for occluded motion estimation" (Short Paper)
- Martin Welk (UMIT TIROL), "Spot and Edge Feature Based Estimation of Point-Spread Functions for Image Deconvolution" (Full Paper)
- Donghwan Lee (Yonsei University), Wooju Kim (Yonsei University), "Organ Level Representation Learning for Region Based Medical Image Retrieval" (Full Paper)
- Christof Kauba (University of Salzburg), Andreas Uhl (University of Salzburg), "Effect of polarization filters on hand vein sample image quality" (Short Paper)
- Olaf Kaehler (JOANNEUM RESEARCH), Werner Bailer (JOANNEUM RESEARCH), Georg Thallinger (JOANNEUM RESEARCH), "Intelligent Augmentation Methods for Training Defect Detection on Circuit Boards" (Full Paper)
- Achraf Guenounou (University of Udine), Jean-Philippe Andreu (JOANNEUM RESEARCH), Harald Ganster (JOANNEUM RESEARCH), "Explainable AI for Efficient Hyperspectral Band Selection in Textile Recycling: A Score-CAM Approach" (Full Paper)
- Jean-Philippe Andreu (JOANNEUM RESEARCH), Maria Jernej (JOANNEUM RESEARCH), Maximilian Klammer (BOKU University), Benjamin Kromoser (BOKU University), "Assessing Compressive Strength of Reclaimed Clay Bricks Using SWIR Hyperspectral Imaging and Deep Learning" (Short Paper)
Important Dates
- Paper submission deadline: 2026-03-14 (extended)
- Notification of acceptance: 2026-03-24
Submission
We solicit full papers (8 pages) of original, unpublished work and application spotlight papers (extended abstracts) (2-4 pages) focusing on the solution of practical problems. Paper lengths exclude references. All submitted papers will undergo a double-blind peer review process by the program committee. Symposium proceedings will be published electronically by TU Wien Library, covering all participating workshops and with a DOI for each paper.
Submit your paper here: CMT (Track AppliedVision)
If you use the CMT link from the CfP, don't forget to select the track Track AppliedVision.
The papers need to be to follow the format given on the AIRoV homepage.