V
Vlad Hosu
Researcher at University of Konstanz
Publications - 39
Citations - 1232
Vlad Hosu is an academic researcher from University of Konstanz. The author has contributed to research in topics: Image quality & Computer science. The author has an hindex of 10, co-authored 34 publications receiving 528 citations. Previous affiliations of Vlad Hosu include Paris Dauphine University.
Papers
More filters
Journal ArticleDOI
KonIQ-10k: An Ecologically Valid Database for Deep Learning of Blind Image Quality Assessment
TL;DR: This work presents a systematic and scalable approach to creating KonIQ-10k, the largest IQA dataset to date, consisting of 10,073 quality scored images, and proposes a novel, deep learning model (KonCept512), to show an excellent generalization beyond the test set.
Proceedings ArticleDOI
KADID-10k: A Large-scale Artificially Distorted IQA Database
TL;DR: It is believed that the annotated set KADID-10k, together with the unlabelled set K ADIS-700k, can enable the full potential of deep learning based IQA methods by means of weakly-supervised learning.
Proceedings ArticleDOI
The Konstanz natural video database (KoNViD-1k)
Vlad Hosu,Franz Hahn,Mohsen Jenadeleh,Hanhe Lin,Hui Men,Tamás Szirányi,Shujun Li,Dietmar Saupe +7 more
TL;DR: KoNViD-1k is reported on, a subjectively annotated VQA database consisting of 1,200 public-domain video sequences, fairly sampled from a large public video dataset, YFCC100m, aimed at ‘in the wild’ authentic distortions.
Proceedings ArticleDOI
Effective Aesthetics Prediction With Multi-Level Spatially Pooled Features
TL;DR: This work proposes the first method that efficiently supports full resolution images as an input, and can be trained on variable input sizes, and significantly improves upon the state of the art on ground-truth mean opinion scores.
Journal ArticleDOI
KonIQ-10k: Towards an ecologically valid and large-scale IQA database
TL;DR: This work shows how it built an IQA database, KonIQ-10k, consisting of 10,073 images, on which it argues for its ecological validity by analyzing the diversity of the dataset, by comparing it to state-of-the-art IQA databases, and by checking the reliability of user studies.