Alireza Ghasemi

Data Scientist at ELCA

You can see my CV in the short or long form.

About Me

I'm a machine learning and computer vision researcher as well as an expert Java developer. I have been actively working in the machine learning, computer vision and related areas since 2008 and I have mostly worked with massive amounts of data since then. I worked with diverse modalities of data including text information in multiple languages (poetry detection in Persian text, statistical machine translation, news search engine, website categorization), speech signals and visual information (video data, light-field images, color photographs). In all these cases, what I did was essentially organizing masses of data and summarizing them into meaningful and desirable information. This is the task I enjoy most: seeing meaningful information appearing out of masses of seemingly incomprehensible data.

Besides research and looking for novel solutions which has led to a track record of scientific publications in well-known international venues, I have industry-level experience in implementing machine learning and data mining solutions and developing products out of them. I have more than 10 years of expertise in Java development developing desktop, web, and also Android applications. As well as these, I have great working knowledge of NoSQL databases as well as in-depth experience with relational engines. I'm also fluent in MATLAB and experienced in a variety of other languages, including Python and JavaScript. I have been programming since high school and I still enjoy it a lot.

Current Research:

The imaging technology is rapidly changing. The so-called light-field or plenoptic cameras can dramatically change the way we capture photos, both for entertainment and commercial purposes. This is mostly because of the vast amount of visual information available in the plenoptic function, compared to traditional photos.

Currently, I'm working on plenoptic image processing and computer vision. I will develop novel and efficient methods and algorithms to analyze and extract features from plenoptic images by making best use of the information inherent in them. Utilizing plenoptic information can significantly improve the accuracy and efficiency of current traditional computer vision algorithms.

Past Research:

For my M. Sc. studies, I worked on applications of machine learning and pattern recognition to image retrieval problems (Link in Persian). Specifically, I studied semi-supervised and active learning methods and their applicability to image retrieval and categorization. I worked on different measures for sample selection in active learning and trying to utilize active learning in improving relevance feedback efficiency in image retrieval. I proposed and implemented a novel one-class learning method which utilizes unlabeled data and can be used as a an active learning approach.

I also studied one-class learning algorithms, support estimation methods and methods for learning with few or imbalanced data because these limitations directly apply to image retrieval problem.

I’ve had experiences in different classification tasks, from document classification to mood classification and poetry detection in text data. I have done my Bachelor’s thesis on website classification which is a special classification task because of presence of links between pages and different page categories. I’ve had also experiments with different term weighting schemes for information retrieval and text classification. In addition to these I’ve worked on prediction of people’s psychological characteristics from their personal blogs’ posts.


September 2011 - December 2016: PhD in Computer, Communication and Information Sciences, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland. Supervised by Professor Martin Vetterli and Dr. Adam James Scholefield.
September 2009 - August 2011: M. Sc. in Computer Engineering - Artificial Intelligence, Sharif University of Technology, Tehran, Iran.
September 2005 - September 2009: B. Sc. in Computer Engineering - Software Engineering, Sharif University of Technology, Tehran, Iran.
September 2001 - July 2005: Diploma in Mathematics and Physics, Shahid Ejei High School (Part of NODET), Esfahan, Iran.

Professional Experience

January 2017 - Present: Data Scientist and Software Engineer, ELCA Informatique, Lausanne, Switzerland.
January 2012 - December 2016: Doctoral Researcher and R&D Engineer at Audiovisual Communications Lab (LCAV), École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland.
October 2015 - March 2016: R&D Intern at Swiss Center for Electronics and Microtechnology (CSEM), Neuchatel, Switzerland.
September 2011 - January 2012: Doctoral Assistant at Artificial Intelligence Lab (LIA), École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland.
March 2010 - August 2011: Research Assistant at DSP Lab, Sharif University of Technology, Tehran, Iran.
October 2009 - March 2010: Web Developer at Intelligent Information Systems (IIS) Lab, Sharif University of Technology, Tehran, Iran.
May 2009 - October 2009: Backend Service Developer at Peykasa Messageware Group, Tehran, Iran.
January 2009 - May 2009: R&D Assistant at Semantic Web Research Lab, Sharif University of Technology, Tehran, Iran.
June 2008 - September 2008: Software Engineering Intern at Mabna Software Co., Tehran, Iran.


You can visit my Google Scholar profile.


Alireza Ghasemi, Adam James Scholefield and Martin Vetterli: Consistent and Optimal Reconstruction of Linearly-Related Variables. Filed September 2015. Demo I Demo II Demo III Demo IV
Alireza Ghasemi, Laurent Rime and Martin Vetterli: Method and Apparatus for Identifying Local Features. US Patent No. 20140369594, Filed June 2013.
Alireza Ghasemi, Laurent Rime, Martin Vetterli, Distinguishing Real Scenes from Printed Photos Using a Light-Field Camera. US-61898739, Provisional Patent Filed November 2013.


Alireza Ghasemi, Adam James Scholefield and Martin Vetterli: SHAPE: Linear-Time Camera Pose Estimation With Quadratic Error-Decay. ICASSP 2016, Shanghai, China. Demo
Alireza Ghasemi, Adam James Scholefield and Martin Vetterli: On the Accuracy of Point Localisation in a Circular Camera-Array. ICIP 2015, Quebec City, Canada. (Chosen in the top 10% papers) Demo
Alireza Ghasemi and Martin Vetterli: Detecting Planar Surface Using a Light-Field Camera with Application to Distinguishing Real Scenes From Printed Photos. ICASSP 2014, Florence, Italy.
Alireza Ghasemi, Mahdad Hosseini Kamal and Martin Vetterli: Computationally Efficient Back- ground Subtraction in the Light Field Domain. IS&T/SPIE Electronic Imaging 2014, San Francisco. California, USA, February 2-6, 2014.
Alireza Ghasemi, Nelly Afonso and Martin Vetterli: LCAV-31: A Dataset for Light Field Object Recognition. IS&T/SPIE Electronic Imaging 2014, San Francisco. California, USA, February 2-6, 2014.
Alireza Ghasemi and Martin Vetterli: Scale-Invariant Representation of Light Field Images for Object Recognition and Tracking. IS&T/SPIE Electronic Imaging 2014, San Francisco. California, USA, February 2-6, 2014.
Claudiu Cristian Musat, Alireza Ghasemi, Boi Faltings: Sentiment Analysis Using a Novel Human Computation Game. ACL 2012 People’s Web Meets NLP Workshop, 2012.
Alireza Ghasemi, Hamid R. Rabiee, Mohammad T. Manzuri, Mohammad H. Rohban: A Bayesian Approach to the Data Description Problem. AAAI Twenty-Sixth Conference on Artificial Intelli- gence (AAAI-12), 2012
Amirhossein Tavanaei, Alireza Ghasemi, Mohammad Tavanaei, Hossein Sameti, Mohammad T. Manzuri: Support Vector Data Description for Spoken Digit Recognition. International Conference on Bio-inspired Systems and Signal Processing (BIOSIGNALS 2012), 2012:32-37.
Alireza Ghasemi, Hamid R. Rabiee, Mohsen Fadaee, Mohammad T. Manzuri, Mohammad H. Rohban: Active Learning from Positive and Unlabeled Data. IEEE ICDM 2011 Workshop on Optimization Based Methods for Emerging Data Mining Problems (OEDM’11), 2011:244-250.
Alireza Ghasemi, Mohammad T. Manzuri, Hamid R. Rabiee, Mohammad H. Rohban, Siavash Haghiri: Active One-Class Learning by Kernel Density Estimation. IEEE International Workshop on Machine Learning for Signal Processing (MLSP’11), 2011:1-6.


Alireza Ghasemi: Feature Identification and Tracking in the Plenoptic Space, Candidacy Report for the Doctoral Program in Computer and Communication Systems, Under Supervision of Prof. Martin Vetterli, Summer 2012.
Alireza Ghasemi: Content-Based Image Retrieval Using Relevance Feedback and Semi-Supervised Learning, M. Sc. Thesis, Under Supervision of Dr. Mohammad Taghi Manzuri, Summer 2011.
Alireza Ghasemi: A Survey on Website Classification Methods with Introduction to a New Method Based on Internal PageRanks, B.Sc. Thesis, Under Supervision of Dr. Hassan Abolhassani, Summer 2009.


Alireza Ghasemi, Julien Lalande, Laurent Rime, Martin Vetterli: Light-Field/Plenoptic Image Processing - Principles. Research Day of the EPFL School of Computer and Communication Sciences, June 2014.
Alireza Ghasemi, Julien Lalande, Laurent Rime, Martin Vetterli: Light-Field/Plenoptic Image Processing - Applications. Research Day of the EPFL School of Computer and Communication Sciences, June 2014.

Honors and Awards

Fall 2015: Granted the EPFL Bonus for Exceptional Performance.
Spring 2015: Granted IEEE Signal Processing Society (SPS) Travel Grant to attend the conference ICIP 2015.
Spring 2015: Granted the Qualcomm Innovation Fellowship 2015.
Fall 2011: Granted full one-year fellowship by the Doctoral School of the I&C department to start PhD at EPFL.
Summer 2011: Ranked 1st by GPA among all Artificial Intelligence students in the Sharif University of Technology.
Spring 2009: Ranked 1st in the nationwide graduate entrance exam in Computer Engineering - Artificial Intelligence among more than 17000 applicants.
Spring 2009: Ranked 1st in the nationwide graduate entrance exam in IT Engineering among more than 13000 applicants.
Summer 2009: Ranked 5th in first round of the nationwide scientific students’ olympiad in Computer Engineering.
Fall 2008: Honorary admission for graduate study in the Sharif University of Technology.
Summer 2005: Ranked 183rd in the nationwide university entrance exam (among more than 300,000 students), Mathematics section.
Summer 2005: Ranked 33rd in nationwide university entrance exam, foreign languages section (among more than 500,000 students).


Java: Expert. Java has been my primary programming language since 2005. I have done many types of projects, from desktop GUIs (Swing, JavaFX) to large-scale application servers to scientific computing (Machine learning using Spark,SMILE, and DL4J, NLP using CoreNLP, Computational Geometry, Computer Vision using OpenCV, $\dots$) and Android development.
Python: I have a broad experience in writing prototypes and PoCs and conducting analyses for machine learning and data analysis tasks using various tools and libraries including TensorFlow, Keras, NumPy/SciPy, NLTK, and XGBoost, as well as automation scripts for software environments such as the Blender 3-D rendering software.
Matlab: Proficient. I have been using Matlab in research since 2007. I have done tens of small and medium-level projects, mostly in the machine learning, signal processing and computer vision area, but also in simulation and optimization.
JavaScript: Experienced. I have used JavaScript in server-side programming (Node.js and Meteor) and Google Chrome extension development as well as client-side scripting.
SQL: Experienced. I used plain SQL and JDBC in small and medium-sized projects from 2008 to 2011. I have experiences in database design (ER diagrams, $\dots$) as well and have taught undergraduate database courses for numerous classes from 2009 to 2011.
C/C++: Experienced. I have been programming in C/C++ since high-school (circa 2002).
PHP, Groovy: I did medium-level projects in PHP and Groovy from 2009 to 2011.
$\LaTeX$: Fluent and experienced. I have been actively using and developing $\LaTeX$ since 2007 for every type of document (Reports, posters, presentation slides,$\dots$).
Web Application Development: I have done small and medium-level projects using various web frameworks including Symfony, Spring, Struts2, GWT, Grails, $\ldots$.
NoSQL: I like NoSQL databases. I have been using Firebase and MongoDB in my projects (Android and client-side JavaScript) since 2014.
Operating Systems: I have worked with various flavours of Linux (Debian, Fedora, Mint, Ubuntu) and Windows for both personal and professional use. I currently administer a virtual private server (VPS) running Debian for personal purposes. I also have experiences with Win32 programming.
Version Control and Build Automation Systems: I have worked with Git and SVN for version controlling in the projects I've been involved. I'm also a user of Maven and Gradle for managing dependencies and build automation in the Java environment.

Language Portfolio

You may want to looke at my Duolingo profile. I happily use Duolingo to learn new languages.

Persian: Native.
English: Professional fluency. I use English intensively in the work and study environments. C1 level.
French: Limited fluency. I've been learning French since 2012. Currently in the B1 level.
German: Limited fluency. I've been learning German since 2013. Currently in the A2 level.
Arabic: Basic. I learned the essentials of the Arabic language in the middle and highschool.