1 code implementation • 9 Apr 2024 • Jiayi Shen, Cheems Wang, Zehao Xiao, Nanne van Noord, Marcel Worring
This paper proposes \textit{GO4Align}, a multi-task optimization approach that tackles task imbalance by explicitly aligning the optimization across tasks.
1 code implementation • 29 Jan 2024 • Carlo Bretti, Pascal Mettes, Hendrik Vincent Koops, Daan Odijk, Nanne van Noord
Creating a trailer requires carefully picking out and piecing together brief enticing moments out of a longer video, making it a chal- lenging and time-consuming task.
no code implementations • 10 Oct 2023 • Alexandra Barancová, Melvin Wevers, Nanne van Noord
This paper explores the capacity of computer vision models to discern temporal information in visual content, focusing specifically on historical photographs.
1 code implementation • 5 Sep 2023 • Nanne van Noord
However, when applied to a single dataset the discovery of visual concepts is restricted to those most prominent.
1 code implementation • 1 Jan 2023 • Sadaf Gulshad, Teng Long, Nanne van Noord
To interpret deep neural networks, one main approach is to dissect the visual input and find the prototypical parts responsible for the classification.
1 code implementation • ICCV 2023 • Nanne van Noord
However, when applied to a single dataset the discovery of visual concepts is restricted to those most prominent.
no code implementations • ICCV 2023 • Teng Long, Nanne van Noord
Our findings demonstrate the strength of Hyperbolic Hierarchical Clustering and its potential for Self-Supervised Learning.
no code implementations • 14 Nov 2022 • Nanne van Noord, Melvin Wevers, Tobias Blanke, Julia Noordegraaf, Marcel Worring
We believe that possible implementations of these aspects into AI research leads to AI that better captures the complexities of culture.
1 code implementation • CVPR 2022 • Mina GhadimiAtigh, Julian Schoep, Erman Acar, Nanne van Noord, Pascal Mettes
For image segmentation, the current standard is to perform pixel-level optimization and inference in Euclidean output embedding spaces through linear hyperplanes.
no code implementations • 3 Feb 2022 • Nikolaos-Antonios Ypsilantis, Noa Garcia, Guangxing Han, Sarah Ibrahimi, Nanne van Noord, Giorgos Tolias
Testing is primarily performed on photos taken by museum guests depicting exhibits, which introduces a distribution shift between training and testing.
1 code implementation • 21 Dec 2021 • Mariya Hendriksen, Maurits Bleeker, Svitlana Vakulenko, Nanne van Noord, Ernst Kuiper, Maarten de Rijke
One aspect of this data is a category tree that is being used in search and recommendation.
1 code implementation • 26 Nov 2021 • Sarah Ibrahimi, Nanne van Noord, Tim Alpherts, Marcel Worring
Additionally, we introduce a new training protocol Inside Out Data Augmentation to adapt Visual Place Recognition methods for localizing indoor images, demonstrating the potential of Inside Out Visual Place Recognition.
2 code implementations • 3 Sep 2019 • Maximilian Müller-Eberstein, Nanne van Noord
The Synesthetic Variational Autoencoder (SynVAE) introduced in this research is able to learn a consistent mapping between visual and auditive sensory modalities in the absence of paired datasets.
no code implementations • 7 Aug 2019 • Laurens Samson, Nanne van Noord, Olaf Booij, Michael Hofmann, Efstratios Gavves, Mohsen Ghafoorian
Adversarial training has been recently employed for realizing structured semantic segmentation, in which the aim is to preserve higher-level scene structural consistencies in dense predictions.
no code implementations • 5 Apr 2019 • Gjorgji Strezoski, Nanne van Noord, Marcel Worring
When task relations are explicitly defined based on domain knowledge multi-task learning (MTL) offers such concurrent solutions, while exploiting relatedness between multiple tasks performed over the same dataset.
1 code implementation • ICCV 2019 • Gjorgji Strezoski, Nanne van Noord, Marcel Worring
Typical multi-task learning (MTL) methods rely on architectural adjustments and a large trainable parameter set to jointly optimize over several tasks.
no code implementations • 17 Jan 2018 • Nanne van Noord, Eric Postma
In this work we propose Pixel Content Encoders (PCE), a light-weight image inpainting model, capable of generating novel con-tent for large missing regions in images.
no code implementations • 3 Feb 2016 • Nanne van Noord, Eric Postma
This leads to the conclusion that encouraging the combined development of a scale-invariant and scale-variant representation in CNNs is beneficial to image recognition performance.
no code implementations • 19 Jun 2015 • Nanne van Noord, Eric Postma
Previous work has shown that the artist of an artwork can be identified by use of computational methods that analyse digital images.