My areas of interest lie in the field of Machine Learning, Data Mining, Artificial Intelligence and Technologies, specifically in Computer Vision and Pattern Recognition. Within this field, my current research is focused on Facial Expressions Recognition and Affective Computing. I also enjoy other related fields involving Deep Learning techniques, Image Processing, Programming Languages and Photography.

I have been awarded three years in a row by the Google Research Awards for Latin America (2017, 2016, 2015), which recently changed its name to Latin America Research Awards (LARA), with my project: “Learning Dynamic Action Units for Three-­dimensional Facial Expression Recognition” guided by Prof. Pablo Arbelaez.

Press coverage:



Andrés Felipe Romero (left) and Pablo Arbelaez (right) were interviewed by one of the most important Colombian News, RCN. Follow the link to see the video.
Andrés Felipe Romero and Pablo Arbelaez’s project is the only project that has been awarded for three years in a row amongst more than 1200 project proposals presented for the Latin America Research Awards (LARA) since 2013.


Andrés Felipe Romero talked with LaFM radiostation about computers that would be able to recognice human emotions and more stuffs.
Andrés Felipe Romero and Winston Percybrooks, the only Colombian winners for the Latin America Research Awards.


Andrés Felipe Romero (center) is part of the 17 selected projects awarded by “Google Research Awards for Latin America 2016”. More than 800 project proposals were presented to the 2016 version.
Andrés Felipe Romero (left) receives award by the hand of Berthier Ribeiro (derecha), who is the Director of Engineering at Google Research and Development Center, at Belo Horizonte (Brazil).


Andrés with his certificate during the opening ceremony of the “Google Research Awards for Latin America 2016” in Belo Horizonte (Brazil).
Andrés Felipe Romero (left) and his proud father, Antonio Romero (right), were at a local newspaper situated at Sincelejo (Sucre), which is his native department.


Andrés Felipe Romero (left) and Pablo Arbelaez (right) are part of the 12 selected projects awarded by “Google Research Awards for Latin America 2015”, being the only Colombian group accomplishing this feat, among 600 project proposals.

Currently, I’m doing a PhD in Engineering at Universidad de los Andes, CO.


  • Abstract: We propose a novel convolutional neural network architecture to address the  fine-grained recognition problem of multi-view dynamic facial action unit detection. We leverage recent gains in large-scale object recognition by formulating the task of predicting the presence or absence of a specific action unit in a still image of a human face as holistic classification. We then explore the design space of our approach by considering both shared and independent representations for separate action units, and also different CNN architectures for combining color and motion information. We then move to the novel setup of the FERA 2017 Challenge, in which we propose a multi-view extension of our approach that operates by first predicting the viewpoint from which the video was taken, and then evaluating an ensemble of action unit detectors that were trained for that specific viewpoint. Our approach is holistic, efficient, and modular, since new action units can be easily included in the overall system. Our approach significantly outperforms the baseline of the FERA 2017 Challenge, which was the previous state-of-the-art in multi-view dynamic action unit detection, with an absolute improvement of 14%.

Research Group

I’m part of the Biomedical Computer Vision (BCV) group at Universidad de los Andes, Bogotá, Colombia. Our lead researcher is Prof Pablo Arbelaez.

Homepage: https://biomedicalcomputervision.uniandes.edu.co/

Meet some of them:

In our group we study several medical and non-medical problems such as: Automated Detection of Lung Cancer with Artificial Intelligence, Detection of Breast Cancer using Thermographic Images, Detection of Diabetic Retinopathy using Artificial Intelligence, Action recognition in Videos, among many others.