1 code implementation • 15 Sep 2023 • Elias Ramzi, Nicolas Audebert, Clément Rambour, André Araujo, Xavier Bitot, Nicolas Thome
It provides an upperbound for rank losses and ensures robust training.
no code implementations • ICCV 2023 • Nikolaos-Antonios Ypsilantis, KaiFeng Chen, Bingyi Cao, Mário Lipovský, Pelin Dogan-Schönberger, Grzegorz Makosa, Boris Bluntschli, Mojtaba Seyedhosseini, Ondřej Chum, André Araujo
In this work, we address the problem of universal image embedding, where a single universal model is trained and used in multiple domains.
1 code implementation • ICCV 2023 • Thomas Mensink, Jasper Uijlings, Lluis Castrejon, Arushi Goel, Felipe Cadar, Howard Zhou, Fei Sha, André Araujo, Vittorio Ferrari
Empirically, we show that our dataset poses a hard challenge for large vision+language models as they perform poorly on our dataset: PaLI [14] is state-of-the-art on OK-VQA [37], yet it only achieves 13. 0% accuracy on our dataset.
1 code implementation • ICCV 2023 • Dror Aiger, André Araujo, Simon Lynen
In this paper, we take a step back from this assumption and propose Constrained Approximate Nearest Neighbors (CANN), a joint solution of k-nearest-neighbors across both the geometry and appearance space using only local features.
no code implementations • 22 Mar 2023 • Arjun Karpur, Guilherme Perrotta, Ricardo Martin-Brualla, Howard Zhou, André Araujo
Finding localized correspondences across different images of the same object is crucial to understand its geometry.
no code implementations • 2 Jun 2022 • Zu Kim, André Araujo, Bingyi Cao, Cam Askew, Jack Sim, Mike Green, N'Mah Fodiatu Yilla, Tobias Weyand
We showcase its application to the landmark recognition domain, presenting a detailed analysis and the final fairer landmark rankings.
no code implementations • 19 Aug 2021 • Zu Kim, André Araujo, Bingyi Cao, Cam Askew, Jack Sim, Mike Green, N'Mah Fodiatu Yilla, Tobias Weyand
To create a more comprehensive and equitable dataset, we start by defining the fair relevance of a landmark to the world population.
3 code implementations • 12 Apr 2021 • Ahmet Iscen, André Araujo, Boqing Gong, Cordelia Schmid
An effective and simple approach to long-tailed visual recognition is to learn feature representations and a classifier separately, with instance and class-balanced sampling, respectively.
Ranked #11 on Long-tail Learning on iNaturalist 2018