The technical program features technical sessions, a plenary lecture, a poster session, and an industry session with presentations of companies.

Location: Grey Hall, Faculty of Electrical Engineering and Computing

 

Monday, 11 November 2024
09:00-09:15 Opening
  Oral Session 1
09:15-09:30

Fine-grained Classes and How to Find Them
Matej Grcić1,2, Artyom Gadetsky1 and Maria Brbić1
1EPFL, Lausanne, Switzerland; 2University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
09:30-09:45 MC-PanDA: Mask Confidence for Panoptic Domain Adaptation (arXiv:2407.14110)
Ivan Martinović, Josip Šarić and Siniša Šegvić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
09:45-10:00 Scalable Hypersphere Embedding For Semantic Metric Learning
Lovre Antonio Budimir, Marko Subašić, Zoran Kalafatić and Sven Lončarić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
10:00-10:15 Extensive evaluation of image classifiers’ interpretations
Suraja Poštić and Marko Subašić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
10:15-10:30 Break
10:30-11:15 Plenary Lecture/IEEE SPS DL Talk
Object-centric Generative AI
Justin Dauwels
Delft University of Technology, The Netherlands
11:15-11:30 Break
  Oral Session 2
11:30-11:45 Pose-independent 3D Anthropometry from Sparse Data
David Bojanić1,2, Stefanie Wuhrer2, Tomislav Petković1 and Tomislav Pribanić1†
1University of Zagreb Faculty of Electrical Engineering and Computing, Croatia; 2 Inria centre at the University Grenoble Alpes, Grenoble, France
11:45-12:00 Modular Transformer-Based Monitoring and Prediction of Wet AMD progression Using OCT Imaging
Lovre Antonio Budimir, Ivana Matovinović, Donik Vršnak and Sven Lončarić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
12:00-12:15 Efficient Method for Robust Backdoor Detection and Removal in Feature Space using Clean Data
Donik Vršnak, Marko Subašić and Sven Lončarić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
12:15-12:30 Reconstruction of incomplete wildfire data using deep generative models
Tomislav Ivek1 and Domagoj Vlah2
1Institute of Physics, Zagreb, Croatia; 2University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
12:30-14:00 Break, Demonstrations and Poster Session
Posters on current computer vision research activities in Croatia
(See list of posters below.)
  Industry session
14:00-14:15 Advanced Technologies for Processing Images from Next-Generation Video Sensors
Roko Končurat
Xylon
14:15-14:30

Enhancing real-time monocular depth estimation using knowledge distillation
Marko Peroš and Bartol Markovinović
Visage Technologies
14:30-14:45 Break
  Presentation of Projects and Laboratories
14:45-15:00 Maritime Open-vocabulary Panoptic Segmentation (MOPS)
Josip Šarić
University of Ljubljana Faculty of Computer and Information Science, Slovenia
15:00-15:15 Project Robust Structured Light Coding for 3D Imaging in Difficult Conditions: On Underwater 3D Scanning
Tomislav Petković
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
15:15-15:30

Razumijevanje prometnih scena za procjenu sigurnosti Vijetnamskih škola
Siniša Šegvić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia
  Closing

 

Poster Session

1. Darts game detection and tracking vision system
Jurica Cvetić and Filip Šuligoj
University of Zagreb Faculty of Mechanical Engineering and Naval Architecture, Croatia

2. Autonomous Quadcopter Navigation for Search and Rescue Missions Using Computer Vision and Convolutional Neural Networks
Luka Šiktar, Filip Šuligoj, Bojan Šekoranja, Marko Švaco and Branimir Ćaran
University of Zagreb Faculty of Mechanical Engineering and Naval Architecture, Croatia

3. Perspective-Based Object-Level Augmentation for Object Detection
Katarina Tolja, Marko Subašić, Sven Lončarić and Zoran Kalafatić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia

4. Features Selection for Autism Spectrum Disorder Classification from Functional MRI
Lucija Petkoviček and Jelena Božek
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia

5. A Survey on Deep Learning-based Gaze Direction Regression: Searching for the State-of-the-art
Franko Šikić, Donik Vršnak and Sven Lončarić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia

6. Instance Detection and Geospatial Registration of Trees in Urban Imagery
Hrvoje Nuić, Željka Mihajlović and Zoran Kalafatić
University of Zagreb Faculty of Electrical Engineering and Computing, Croatia