no code implementations • 17 Mar 2024 • Boujemaa Guermazi, Riadh Ksantini, Naimul Khan
This work presents an improved version of an unsupervised Convolutional Neural Network (CNN) based algorithm that uses a constant weight factor to balance between the segmentation criteria of feature similarity and spatial continuity, and it requires continuous manual adjustment of parameters depending on the degree of detail in the image and the dataset.
1 code implementation • 28 Dec 2023 • Nairouz Mrabah, Mohamed Bouguessa, Riadh Ksantini
Thanks to a newly identified term, our lower bound can escape Posterior Collapse and has more flexibility to account for the difference between the inference and generative models.
no code implementations • 29 Sep 2022 • Mohamed Amine Ouali, Mohamed Bouguessa, Riadh Ksantini
We devise a graph attention network-based approach for learning a scene triangle mesh representation in order to estimate an image camera position in a dynamic environment.
1 code implementation • 19 Jul 2021 • Nairouz Mrabah, Mohamed Bouguessa, Mohamed Fawzi Touati, Riadh Ksantini
We study these issues from two aspects: (1) there is a trade-off between Feature Randomness and Feature Drift when clustering and reconstruction are performed at the same level, and (2) the problem of Feature Drift is more pronounced for GAE models, compared with vanilla auto-encoder models, due to the graph convolutional operation and the graph decoding design.
Ranked #1 on Node Clustering on Pubmed
1 code implementation • 23 Dec 2020 • Ahmed Zgaren, Wassim Bouachir, Riadh Ksantini
Motivated by this observation, and by the fact that discriminative correlation filters(DCFs) may provide a complimentary low-level information, we presenta novel tracking algorithm taking advantage of both approaches.
no code implementations • arXiv 2019 • Nairouz Mrabah, Mohamed Bouguessa, Riadh Ksantini
Clustering using deep autoencoders has been thoroughly investigated in recent years.
Ranked #2 on Image Clustering on MNIST-full
1 code implementation • 23 Jan 2019 • Nairouz Mrabah, Naimul Mefraz Khan, Riadh Ksantini, Zied Lachiri
In unsupervised learning, there is no apparent straightforward cost function that can capture the significant factors of variations and similarities.
Ranked #1 on Image Clustering on MNIST-test
no code implementations • 15 Oct 2016 • Takoua Kefi, Riadh Ksantini, M. Becha Kaaniche, Adel Bouhoula
This research contributes to the field of one-class incremental learning and classification in case of non-stationary environments.