no code implementations • 23 Apr 2024 • Merey Ramazanova, Alejandro Pardo, Bernard Ghanem, Motasem Alfarra
Understanding videos that contain multiple modalities is crucial, especially in egocentric videos, where combining various sensory inputs significantly improves tasks like action recognition and moment localization.
1 code implementation • 10 Apr 2023 • Motasem Alfarra, Hani Itani, Alejandro Pardo, Shyma Alhuwaider, Merey Ramazanova, Juan C. Pérez, Zhipeng Cai, Matthias Müller, Bernard Ghanem
To address this issue, we propose a more realistic evaluation protocol for TTA methods, where data is received in an online fashion from a constant-speed data stream, thereby accounting for the method's adaptation speed.
1 code implementation • 3 Apr 2023 • Joachim Houyon, Anthony Cioppa, Yasir Ghunaim, Motasem Alfarra, Anaïs Halin, Maxim Henry, Bernard Ghanem, Marc Van Droogenbroeck
In this paper, we propose a solution to this issue by leveraging the power of continual learning methods to reduce the impact of domain shifts.
1 code implementation • CVPR 2023 • Yasir Ghunaim, Adel Bibi, Kumail Alhamoud, Motasem Alfarra, Hasan Abed Al Kader Hammoud, Ameya Prabhu, Philip H. S. Torr, Bernard Ghanem
We show that a simple baseline outperforms state-of-the-art CL methods under this evaluation, questioning the applicability of existing methods in realistic settings.
no code implementations • CVPR 2023 • Andrés Villa, Juan León Alcázar, Motasem Alfarra, Kumail Alhamoud, Julio Hurtado, Fabian Caba Heilbron, Alvaro Soto, Bernard Ghanem
In this paper, we address the problem of continual learning for video data.
no code implementations • 29 Nov 2022 • Motasem Alfarra, Zhipeng Cai, Adel Bibi, Bernard Ghanem, Matthias Müller
This work explores the problem of Online Domain-Incremental Continual Segmentation (ODICS), where the model is continually trained over batches of densely labeled images from different domains, with limited computation and no information about the task boundaries.
no code implementations • 29 Sep 2022 • Kumail Alhamoud, Hasan Abed Al Kader Hammoud, Motasem Alfarra, Bernard Ghanem
Recent progress in empirical and certified robustness promises to deliver reliable and deployable Deep Neural Networks (DNNs).
1 code implementation • 6 Jun 2022 • Motasem Alfarra, Juan C. Pérez, Egor Shulgin, Peter Richtárik, Bernard Ghanem
However, as in the single-node supervised learning setup, models trained in federated learning suffer from vulnerability to imperceptible input transformations known as adversarial attacks, questioning their deployment in security-related applications.
1 code implementation • CVPR 2022 • Gabriel Pérez S., Juan C. Pérez, Motasem Alfarra, Silvio Giancola, Bernard Ghanem
In this work, we propose 3DeformRS, a method to certify the robustness of point cloud Deep Neural Networks (DNNs) against real-world deformations.
no code implementations • 10 Feb 2022 • Juan C. Pérez, Motasem Alfarra, Ali Thabet, Pablo Arbeláez, Bernard Ghanem
We propose a methodology for assessing and characterizing the robustness of FRMs against semantic perturbations to their input.
1 code implementation • 31 Jan 2022 • Motasem Alfarra, Juan C. Pérez, Anna Frühstück, Philip H. S. Torr, Peter Wonka, Bernard Ghanem
Finally, we show that the FID can be robustified by simply replacing the standard Inception with a robust Inception.
1 code implementation • 29 Jul 2021 • Juan C. Pérez, Motasem Alfarra, Guillaume Jeanneret, Laura Rueda, Ali Thabet, Bernard Ghanem, Pablo Arbeláez
Deep learning models are prone to being fooled by imperceptible perturbations known as adversarial attacks.
1 code implementation • 9 Jul 2021 • Francisco Eiras, Motasem Alfarra, M. Pawan Kumar, Philip H. S. Torr, Puneet K. Dokania, Bernard Ghanem, Adel Bibi
Randomized smoothing has recently emerged as an effective tool that enables certification of deep neural network classifiers at scale.
2 code implementations • 2 Jul 2021 • Motasem Alfarra, Adel Bibi, Naeemullah Khan, Philip H. S. Torr, Bernard Ghanem
Deep neural networks are vulnerable to input deformations in the form of vector fields of pixel displacements and to other parameterized geometric deformations e. g. translations, rotations, etc.
1 code implementation • ICML Workshop AML 2021 • Motasem Alfarra, Juan C. Pérez, Ali Thabet, Adel Bibi, Philip H. S. Torr, Bernard Ghanem
Deep neural networks are vulnerable to small input perturbations known as adversarial attacks.
no code implementations • 1 Jan 2021 • Motasem Alfarra, Adel Bibi, Hasan Abed Al Kader Hammoud, Mohamed Gaafar, Bernard Ghanem
This work tackles the problem of characterizing and understanding the decision boundaries of neural networks with piecewise linear non-linearity activations.
no code implementations • 8 Dec 2020 • Motasem Alfarra, Adel Bibi, Philip H. S. Torr, Bernard Ghanem
In this work, we revisit Gaussian randomized smoothing and show that the variance of the Gaussian distribution can be optimized at each input so as to maximize the certification radius for the construction of the smooth classifier.
1 code implementation • 13 Jun 2020 • Motasem Alfarra, Juan C. Pérez, Adel Bibi, Ali Thabet, Pablo Arbeláez, Bernard Ghanem
This paper studies how encouraging semantically-aligned features during deep neural network training can increase network robustness.
no code implementations • 3 May 2020 • Motasem Alfarra, Slavomir Hanzely, Alyazeed Albasyoni, Bernard Ghanem, Peter Richtarik
Recent advances in the theoretical understanding of SGD led to a formula for the optimal batch size minimizing the number of effective data passes, i. e., the number of iterations times the batch size.
no code implementations • 20 Feb 2020 • Motasem Alfarra, Adel Bibi, Hasan Hammoud, Mohamed Gaafar, Bernard Ghanem
Our main finding is that the decision boundaries are a subset of a tropical hypersurface, which is intimately related to a polytope formed by the convex hull of two zonotopes.
1 code implementation • ECCV 2020 • Juan C. Pérez, Motasem Alfarra, Guillaume Jeanneret, Adel Bibi, Ali Thabet, Bernard Ghanem, Pablo Arbeláez
We revisit the benefits of merging classical vision concepts with deep learning models.
no code implementations • 25 Sep 2019 • Motasem Alfarra, Adel Bibi, Hasan Hammoud, Mohamed Gaafar, Bernard Ghanem
We use tropical geometry, a new development in the area of algebraic geometry, to provide a characterization of the decision boundaries of a simple neural network of the form (Affine, ReLU, Affine).