arXiv subjects

Djamila Aouada

Publications and source records attributed to Djamila Aouada.

75 records · Page 5Linked to original sources

arXiv2019

A survey on Deep Learning Advances on Different 3D Data Representations

3D data is a valuable asset the computer vision filed as it provides rich information about the full geometry of sensed objects and scenes. Recently, with the availability of both large 3D datasets and computational power, it is today possible to consider applying deep learning to learn specific tasks on 3D data such as segmentation, recognition and correspondence. Depending on the considered 3D data representation, different challenges may be foreseen in using existent deep learning architectures. In this work, we provide a comprehensive overview about various 3D data representations highlighting the difference between Euclidean and non-Euclidean ones. We also discuss how Deep Learning methods are applied on each representation, analyzing the challenges to overcome.

cs.CV↗

arXiv2015

Ensemble of Example-Dependent Cost-Sensitive Decision Trees

Several real-world classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples and not only within classes. However, standard classification methods do not take these costs into account, and assume a constant cost of misclassification errors. In previous works, some methods that take into account the financial costs into the training of different algorithms have been proposed, with the example-dependent cost-sensitive decision tree algorithm being the one that gives the highest savings. In this paper we propose a new framework of ensembles of example-dependent cost-sensitive decision-trees. The framework consists in creating different example-dependent cost-sensitive decision trees on random subsamples of the training set, and then combining them using three different combination approaches. Moreover, we propose two new cost-sensitive combination approaches; cost-sensitive weighted voting and cost-sensitive stacking, the latter being based on the cost-sensitive logistic regression method. Finally, using five different databases, from four real-world applications: credit card fraud detection, churn modeling, credit scoring and direct marketing, we evaluate the proposed method against state-of-the-art example-dependent cost-sensitive techniques, namely, cost-proportionate sampling, Bayes minimum risk and cost-sensitive decision trees. The results show that the proposed algorithms have better results for all databases, in the sense of higher savings.

cs.LG↗

arXiv2011

Who clicks there!: Anonymizing the photographer in a camera saturated society

In recent years, social media has played an increasingly important role in reporting world events. The publication of crowd-sourced photographs and videos in near real-time is one of the reasons behind the high impact. However, the use of a camera can draw the photographer into a situation of conflict. Examples include the use of cameras by regulators collecting evidence of Mafia operations; citizens collecting evidence of corruption at a public service outlet; and political dissidents protesting at public rallies. In all these cases, the published images contain fairly unambiguous clues about the location of the photographer (scene viewpoint information). In the presence of adversary operated cameras, it can be easy to identify the photographer by also combining leaked information from the photographs themselves. We call this the camera location detection attack. We propose and review defense techniques against such attacks. Defenses such as image obfuscation techniques do not protect camera-location information; current anonymous publication technologies do not help either. However, the use of view synthesis algorithms could be a promising step in the direction of providing probabilistic privacy guarantees.

cs.CR↗