1 code implementation • 27 Mar 2023 • Petra Števuliáková, Petr Hurtik
The main principle is adding a smoothing part to the original IoU, where the smoothing part is given by a linear space with values that increases from the ground truth bounding box to the border of the input image, and thus covers the whole spatial search space.
no code implementations • 24 Mar 2023 • Petr Dvořáček, Petr Hurtik, Petra Števuliáková
We propose a new, simple framework for crafting adversarial examples for black box attacks.
no code implementations • 3 Mar 2023 • Vojtech Molek, Petr Hurtik, Pavel Vlasanek, David Adamczyk
Data shift is a gap between data distribution used for training and data distribution encountered in the real-world.
no code implementations • 10 Apr 2021 • Petr Hurtik, Marek Vajgl
The continuous monitoring of a shore plays an essential role in designing strategies for shore protection against erosion.
no code implementations • 1 Jan 2021 • Petr Hurtik, Marek Vajgl
We present a conceptually simple yet powerful and flexible scheme for refining predictions of bounding boxes.
2 code implementations • 27 May 2020 • Petr Hurtik, Vojtech Molek, Jan Hula, Marek Vajgl, Pavel Vlasanek, Tomas Nejezchleba
We present a new version of YOLO with better performance and extended with instance segmentation called Poly-YOLO.