Petra Bevandić

Dr. Petra Bevandić

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 October 2022
Verified
Book Chapter
Closed access
PDF file
MGMatej Grcic
Matej Grcic
Petra BevandićPetra Bevandić
Petra Bevandić
SŠSiniša Šegvić
Siniša Šegvić

25 October 2019
Verified
Book Chapter
Closed access
PDF file
Petra BevandićPetra Bevandić
Petra Bevandić
IKIvan Krešo
Ivan Krešo
MOMarin Oršić
Marin Oršić
SŠSiniša Šegvić
Siniša Šegvić

15 February 2024 in Sensors
Verified
Journal Article
Open access
PDF file
MGMatej Grcic
Matej Grcic
Petra BevandićPetra Bevandić
Petra Bevandić
ZKZoran Kalafatić
Zoran Kalafatić
SŠSiniša Šegvić
Siniša Šegvić
Standard machine learning is unable to accommodate inputs which do not belong to the training distribution. The resulting models often give rise to confident incorrect predictions which may lead to devastating consequences. This problem is especially demanding in the context of dense prediction since input images may be only partially anomalous. Previous work has addressed dense out-of-distribution detection by discriminative training with respect to off-the-shelf negative datasets. However, real negative data may lead to over-optimistic evaluation due to possible overlap with test anomalies. To this end, we extend this approach by generating synthetic negative patches along the border of the inlier manifold. We leverage a jointly trained normalizing flow due to a coverage-oriented learning objective and the capability to generate samples at different resolutions. We detect anomalies according to a principled information-theoretic criterion which can be consistently applied through training and inference. The resulting models set the new state of the art on benchmarks for out-of-distribution detection in road-driving scenes and remote sensing imagery despite minimal computational overhead.

22 January 2019 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
ISIvan Sikiric
Ivan Sikiric
KBKarla Brkic
Karla Brkic
Petra BevandićPetra Bevandić
Petra Bevandić
IKIvan Kreso
Ivan Kreso
JKJosip Krapac
Josip Krapac
SŠSiniša Šegvić
Siniša Šegvić
Visual cues can be used alongside GPS positioning and digital maps to improve understanding of vehicle environment in fleet management systems. Such systems are limited both in terms of bandwidth and storage space, so minimizing the size of transmitted and stored visual data is a priority. In this paper, we present efficient strategies for computing very short image representations suitable for classifying various types of traffic scenes in fleet management systems. We anticipate that the set of interesting classes will change over time, so we consider image representations that can be trained without knowing the labels of the target dataset. We empirically evaluate and compare the presented methods on a contributed dataset of 11447 labeled traffic scenes. Our results indicate that excellent classification results can be achieved with very short image representations, and that fine-tuning on the target dataset image data is not mandatory. Image descriptors can be as short as 128 components while still offering good performance, even in presence of adverse weather or illumination conditions.

30 January 2024 in International Journal of Computer Vision
Verified
Journal Article
Closed access
PDF file
Petra BevandićPetra Bevandić
Petra Bevandić
MOMarin Oršić
Marin Oršić
JŠJosip Šarić
Josip Šarić
IGIvan Grubišić
Ivan Grubišić
SŠSiniša Šegvić
Siniša Šegvić

12 March 2024 in arXiv
Verified
Preprint
Open access
PDF file
Petra BevandićPetra Bevandić
Petra Bevandić
MOMarin Oršić
Marin Oršić
IGIvan Grubišić
Ivan Grubišić
JŠJosip Šarić
Josip Šarić
SŠSiniša Šegvić
Siniša Šegvić
Deep supervised models have an unprecedented capacity to absorb large quantities of training data. Hence, training on multiple datasets becomes a method of choice towards strong generalization in usual scenes and graceful performance degradation in edge cases. Unfortunately, different datasets often have incompatible labels. For instance, the Cityscapes road class subsumes all driving surfaces, while Vistas defines separate classes for road markings, manholes etc. Furthermore, many datasets have overlapping labels. For instance, pickups are labeled as trucks in VIPER, cars in Vistas, and vans in ADE20k. We address this challenge by considering labels as unions of universal visual concepts. This allows seamless and principled learning on multi-domain dataset collections without requiring any relabeling effort. Our method achieves competitive within-dataset and cross-dataset generalization, as well as ability to learn visual concepts which are not separately labeled in any of the training datasets. Experiments reveal competitive or state-of-the-art performance on two multi-domain dataset collections and on the WildDash 2 benchmark.