Siniša Šegvić

About Siniša Šegvić

Siniša Šegvić, With an exceptional h-index of 25 and a recent h-index of 16 (since 2020), a distinguished researcher at Sveucilište u Zagrebu, specializes in the field of Computer Vision, Pattern Recognition, Machine Learning, Artificial Intelligence, Image Processing.

His recent articles reflect a diverse array of research interests and contributions to the field:

Dense Out-of-Distribution Detection by Robust Learning on Synthetic Negative Data

Hybrid open-set segmentation with synthetic negative data

Quantile-Based Maximum Likelihood Training for Outlier Detection

Identifying label errors in object detection datasets by loss inspection

Outlier detection by ensembling uncertainty with negative objectness

Weakly supervised training of universal visual concepts for multi-domain semantic segmentation

Joint forecasting of feature and feature motion

Revisiting consistency for semi-supervised semantic segmentation

Siniša Šegvić Information

University

Position

Professor of Computer Science

Citations(all)

2012

Citations(since 2020)

1308

Cited By

1009

hIndex(all)

25

hIndex(since 2020)

16

i10Index(all)

41

i10Index(since 2020)

27

Email

University Profile Page

Google Scholar

Siniša Šegvić Skills & Research Interests

Computer Vision

Pattern Recognition

Machine Learning

Artificial Intelligence

Image Processing

Top articles of Siniša Šegvić

Title

Journal

Author(s)

Publication Date

Dense Out-of-Distribution Detection by Robust Learning on Synthetic Negative Data

Sensors

Matej Grcić

Petra Bevandić

Zoran Kalafatić

Siniša Šegvić

2024/1

Hybrid open-set segmentation with synthetic negative data

IEEE transactions on pattern analysis and machine intelligence

Matej Grcić

Siniša Šegvić

2024/4/10

Quantile-Based Maximum Likelihood Training for Outlier Detection

Proceedings of the AAAI Conference on Artificial Intelligence

Masoud Taghikhah

Nishant Kumar

Siniša Šegvić

Abouzar Eslami

Stefan Gumhold

2024/3/24

Identifying label errors in object detection datasets by loss inspection

Marius Schubert

Tobias Riedlinger

Karsten Kahl

Daniel Kröll

Sebastian Schoenen

...

2024

Outlier detection by ensembling uncertainty with negative objectness

arXiv preprint arXiv:2402.15374

Anja Delić

Matej Grcić

Siniša Šegvić

2024/2/23

Weakly supervised training of universal visual concepts for multi-domain semantic segmentation

International Journal of Computer Vision

Petra Bevandić

Marin Oršić

Josip Šarić

Ivan Grubišić

Siniša Šegvić

2024/1/30

Joint forecasting of feature and feature motion

2024/1/9

Revisiting consistency for semi-supervised semantic segmentation

Sensors

Ivan Grubišić

Marin Oršić

Siniša Šegvić

2023/1/13

Plan upravljanja istraživačkim podacima projekta ADEPT

Luka Juras

Andrea Vranić

2023

On advantages of mask-level recognition for outlier-aware segmentation

Matej Grcić

Josip Šarić

Siniša Šegvić

2023

Mitigating backdoor attacks with generative modelling and dataset relabelling

Ivan Sabolic

Ivan Grubišić

Siniša Šegvić

2023/10/13

Normalizing flow based feature synthesis for outlier-aware object detection

Nishant Kumar

Siniša Šegvić

Abouzar Eslami

Stefan Gumhold

2023

Real time dense anomaly detection by learning on synthetic negative data

arXiv preprint arXiv:2305.15227

Anja Delić

Matej Grcić

Siniša Šegvić

2023/5/24

Panoptic SwiftNet: Pyramidal Fusion for Real-Time Panoptic Segmentation

Remote sensing

Josip Šarić

Marin Oršić

Siniša Šegvić

2023/4/7

Dynamic loss balancing and sequential enhancement for road-safety assessment and traffic scene classification

arXiv preprint arXiv:2211.04165

Marin Kačan

Marko Ševrović

Siniša Šegvić

2022/11/8

Applications of generative approaches for artificial intelligence

Siniša Šegvić

2022

Densehybrid: Hybrid anomaly detection for dense open-set recognition

Matej Grcić

Petra Bevandić

Siniša Šegvić

2022/10/20

Multi-domain semantic segmentation with overlapping labels

Petra Bevandić

Marin Oršić

Ivan Grubišić

Josip Šarić

Siniša Šegvić

2022

Dense open-set recognition based on training with noisy negative images

Image and vision computing

Petra Bevandić

Ivan Krešo

Marin Oršić

Siniša Šegvić

2022/8/1

Automatic universal taxonomies for multi-domain semantic segmentation

arXiv preprint arXiv:2207.08445

Petra Bevandić

Siniša Šegvić

2022/7/18

See List of Professors in Siniša Šegvić University(Sveucilište u Zagrebu)

Co-Authors

academic-engine