Emilio J. Almazan

Lead Research Scientist

About

Emilio J. Almazan

Lead Research Scientist

Emilio J. Almazan received the PhD degree in Computer Vision from Kingston University (London, UK) under Graeme Jones in 2014.

He is a lead research scientist in NIQ, working on Natural Language Processing and Machine Learning. Prior, he was a postdoc at York University (Toronto, Canada) where he was advised by James Elder.

His current line of research in NIQ focuses on the generation of Foundational models adapted for the retail domain. Additionally, he serves as advisor for other research initiatives such as product matching and categorization of products using AI technology.

Publications

At NIQ

LEA: Improving sentence similarity robustness to typos using lexical attention bias

Mario Almagro, David Jiménez, Diego Ortego, Emilio J. Almazán

click

Block-SCL: blocking matters for supervised contrastive learning in product matching

Mario Almagro, David Jiménez, Diego Ortego, Emilio J. Almazán, Eva Martínez

click

Survival loss: A neuron death regularizer

Emilio Almazan, Javier Tovar and Alejandro de la Calle

click

Geometric interpretation of a CNN’s last layer

Alejandro de la Calle, Javier Tovar, Emilio J. Almazán, Aitor Aller

click

Deep Learning of Visual and Textual Data for Region Detection Applied to Item Coding

Roberto Arroyo, Javier Tovar, Francisco J. Delgado, Emilio J. Almazán, Diego G. Serrador, Antonio Hurtado

click

Integration of Text-maps in Convolutional Neural Networks for Region Detection among Different Textual Categories

Roberto Arroyo, Javier Tovar, Francisco J. Delgado, Emilio J. Almazán, Diego G. Serrador, Antonio Hurtado

click
Other

“Mcmlsd: A dynamic programming approach to line segment detection.”
Almazan, Emilio J., Ron Tal, Yiming Qian, and James H. Elder.
In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2031-2039. 2017.

“Evaluating features and classifiers for road weather condition analysis.”
Qian, Yiming, Emilio J. Almazan, and James H. Elder.
In 2016 IEEE International Conference on Image Processing (ICIP), pp. 4403-4407. IEEE, 2016.

“Road segmentation for classification of road weather conditions.”
Almazan, Emilio J., Yiming Qian, and James H. Elder.
In Computer Vision–ECCV 2016 Workshops: Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part I 14, pp. 96-108. Springer International Publishing, 2016.

“Investigation of tracking processes applicable to adjacent non-overlapping RGB-D sensors.”
Almazán, Emilio J.
PhD diss., Kingston University, 2014.

“Multiple Non-Overlapping RGB-D Sensors for Tracking People.”
Almazán, Emilio J., and Graeme A. Jones.
Robot. Sci. Syst (2013).

“Tracking people across multiple non-overlapping RGB-D sensors.”
Almazan, Emilio, and Graeme Jones.
In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 831-837. 2013.

Emilio J. Almazan

About

About

Emilio J. Almazan

Publications

Publications

LEA: Improving sentence similarity robustness to typos using lexical attention bias

Block-SCL: blocking matters for supervised contrastive learning in product matching

Survival loss: A neuron death regularizer

Geometric interpretation of a CNN’s last layer

Deep Learning of Visual and Textual Data for Region Detection Applied to Item Coding

Integration of Text-maps in Convolutional Neural Networks for Region Detection among Different Textual Categories