activity
20172021
most citedHDR image reconstruction from a single exposure using deep CNNs

617 citations · 714 across the 7 of their papers we have counts for

collaborators
Showing cs.CVShow all

8 papers · 1 filter

cs.CV2021

How to cheat with metrics in single-image HDR reconstruction

Gabriel Eilertsen, Saghi Hajisharif, Param Hanji +3

Single-image high dynamic range (SI-HDR) reconstruction has recently emerged as a problem well-suited for deep learning methods. Each successive technique demonstrates an improveme…

cs.CV20212 cited

Ensembles of GANs for synthetic training data generation

Gabriel Eilertsen, Apostolia Tsirikoglou, Claes Lundström +1

Insufficient training data is a major bottleneck for most deep learning practices, not least in medical imaging where data is difficult to collect and publicly available datasets a…

cs.CV20206 cited

Classifying the classifier: dissecting the weight space of neural networks

Gabriel Eilertsen, Daniel Jönsson, Timo Ropinski +2

This paper presents an empirical study on the weights of neural networks, where we interpret each model as a point in a high-dimensional space -- the neural weight space. To explor…

cs.CV201947 cited

A Closer Look at Domain Shift for Deep Learning in Histopathology

Karin Stacke, Gabriel Eilertsen, Jonas Unger +1

Domain shift is a significant problem in histopathology. There can be large differences in data characteristics of whole-slide images between medical centers and scanners, making g…

cs.CV2019

Single-frame Regularization for Temporally Stable CNNs

Gabriel Eilertsen, Rafał K. Mantiuk, Jonas Unger

Convolutional neural networks (CNNs) can model complicated non-linear relations between images. However, they are notoriously sensitive to small changes in the input. Most CNNs tra…

cs.CV2018

Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsing

Magnus Wrenninge, Jonas Unger

We introduce Synscapes -- a synthetic dataset for street scene parsing created using photorealistic rendering techniques, and show state-of-the-art results for training and validat…