2 code implementations • 13 Apr 2021 • Jakob Božič, Domen Tabernik, Danijel Skočaj
We also show that mixed supervision with only a handful of fully annotated samples added to weakly labelled training images can result in performance comparable to the fully supervised model's performance but at a significantly lower annotation cost.
Ranked #1 on Defect Detection on KolektorSDD2
1 code implementation • 15 Jul 2020 • Jakob Božič, Domen Tabernik, Danijel Skočaj
We demonstrate the performance of the end-to-end training scheme and the proposed extensions on three defect detection datasets - DAGM, KolektorSDD and Severstal Steel defect dataset - where we show state-of-the-art results.
Ranked #1 on Defect Detection on DAGM2007 (Average Precision metric)
1 code implementation • 1 Apr 2019 • Domen Tabernik, Danijel Skočaj
Automatic detection and recognition of traffic signs plays a crucial role in management of the traffic-sign inventory.
5 code implementations • 20 Mar 2019 • Domen Tabernik, Samo Šela, Jure Skvarč, Danijel Skočaj
This paper presents a segmentation-based deep-learning architecture that is designed for the detection and segmentation of surface anomalies and is demonstrated on a specific domain of surface-crack detection.
Ranked #3 on Defect Detection on KolektorSDD
3 code implementations • 20 Feb 2019 • Domen Tabernik, Matej Kristan, Aleš Leonardis
Convolutional neural networks excel in a number of computer vision tasks.
2 code implementations • CVPR 2018 • Domen Tabernik, Matej Kristan, Aleš Leonardis
Classical deep convolutional networks increase receptive field size by either gradual resolution reduction or application of hand-crafted dilated convolutions to prevent increase in the number of parameters.
no code implementations • 13 Sep 2016 • Domen Tabernik, Matej Kristan, Jeremy L. Wyatt, Aleš Leonardis
We propose a novel analytic model of a basic unit in a layered hierarchical model with both explicit compositional structure and a well-defined discriminative cost function.