3rd Summer School

on Deep Learning for Computer Vision


UPC ETSETB TelecomBCN (June 28 - July 4, 2018)

Deep learning technologies are at the core of the current revolution in artificial intelligence for multimedia data analysis. The convergence of large-scale annotated datasets and affordable GPU hardware has allowed the training of neural networks for data analysis tasks which were previously addressed with hand-crafted features. Architectures such as convolutional neural networks, recurrent neural networks and Q-nets for reinforcement learning have shaped a brand new scenario in signal processing. This course will cover the basic principles and applications of deep learning to computer vision problems, such as image classification, object detection or image captioning.


Open Lectures by Guest Speakers

This Summer School will include two open talks from on Aula Master of A3 building in Campus Nord UPC (directions). This session will be open to the general public.

Dr Kevin McGuinness is an Assistant Professor and Science Foundation Ireland Funded Starting Investigator with the School of Electronic Engineering in Dublin City University. He also works very closely with the Insight Centre for Data Analytics. He finished my B.Sc (Hons) in Computer Applications and Software Engineering in Dublin City University in 2005 and was awarded a Ph.D. from the School of Electronic Engineering in 2009. He has since been a postdoctoral researcher at the CLARITY Centre for Sensor Web Technologies, and a research fellow at the Insight Centre for Data Analytics. His primary research interests are computer vision, deep learning, image and video segmentation, segmentation evaluation, machine learning, content-based multimedia information retrieval, and human-computer interaction.

Talk - Crowd counting and analysis: Understanding images containing medium to large groups of people is becoming an increasingly important application of computer vision. Stadiums, airports, concerts, and cities, routinely facilitate large crowds and need to handle the corresponding safety and logistical issues that arise. Crowded images bring their own particular challenges and many approaches have been proposed for counting, density estimation, behaviour classification, and anomaly detection. This talk will focus on our recent work in Insight/DCU on crowded scene analysis and on how we use deep learning, multi-task learning, and domain adaptation for counting and analysing crowds. I will also describe our 2018 CVPR paper that describes how counting models can be adapted across domains, which allows counting of penguins, cells, cars, and people using a model with 95% shared parameters.

Prof. Laura Leal-Taixé is leading the Dynamic Vision and Learning group at the Technical University of Munich, Germany. She received her Bachelor and Master degrees in Telecommunications Engineering from the Technical University of Catalonia (UPC), Barcelona. She did her Master Thesis at Northeastern University, Boston, USA and received her PhD degree (Dr.-Ing.) from the Leibniz University Hannover, Germany. During her PhD she did a one-year visit at the Vision Lab at the University of Michigan, USA. She also spent two years as a postdoc at the Institute of Geodesy and Photogrammetry of ETH Zurich, Switzerland and one year at the Technical University of Munich. In 2017, she won the Sofja Kovalevskaja Award of 1.65 million euros from the presitgious Humboldt Foundation for her project “socialMaps”. Her research interests are dynamic scene understanding, in particular multiple object tracking and segmentation, as well as machine learning for video analysis.


Thursday, June 28 @ D5-010

Friday, June 29 @ D5-010

Monday, July 2 @ D5-010

Tuesday, July 3 @ D5-010

Wednesday, July 4 @ A3-Aula Master (Open Day, register here)


Check the pictures from the 2017 edition.



  • Course codes: 230360 (Master)
  • ECTS credits: 3 (corresponds to full-time dedication during the week course)
  • Teaching language: English
  • The course is offered for both master and bachelor students, but under two study programmes adapted to each profile.
  • Class Dates: June 28 - July 4, 2018
  • Class Schedule: 10am-2pm (you will need 4 extra hours a day for homework during the week course)
  • Capacity: 40 students
  • Location: Campus Nord UPC, Module D5, Room 010


This Summer School requires a previous knowledge on basic deep learning techniques, which will not be covered. Please follow these indications depending on your profile:

If you have no previous experience on deep learning:

You should not sign up for this course. Build a strong basis whether by attending to our next BSc Winter School on Introduction to Deep Learning, or signing up for the full master course of Deep Learning for Artificial Intelligence to be taught during Autumn 2018.

If you have taken a previous edition of DLAI, DLCV, DLSL or have previous experience on deep learning:

  • Master students at ETSETB: Registration is available from the ETSETB academic office. There is an extraordinary registration between May 28th and June 1st, 2018. You can request it from esecretaria and choose “Procedures > related to enrollment”. There you must choose “Enrollment change” or “add a subject”. Only requests made this way will be taken into account.

  • Bachelor (grau) students at ETSETB: You can register these credits as “conjunt d’activitats d’extensió universitària”. Registration is available from the ETSETB academic office. There is an extraordinary registration between May 28th and June 1st, 2018. You can request it from esecretaria and choose “Procedures > related to enrollment”. There you must choose “Enrollment change” or “add a subject”. Only requests made this way will be taken into account.

  • CFIS students: Register directly at the CFIS academic office.

  • Master students at FIB: Contact the FIB academic office. They will collect all applications and submit them to ETSETB for approval.

  • Other students at UPC & Master in Computer Vision: You might audit the course, with no official certification. If interested, fill in this form before 14 June 2018. You may be granted a seat, if there are any available after ETSETB, CFIS & FIB students have signed up.

  • Local industry members and other students from the European Higher Education Area: You must apply for being accepted in the course and cover the 100% cost of the ECTS credits, without the support of the public funds. This corresponds to 143,08 € per ECTS credit (Summer 2016). You may have a seat, if there are any available after UPC students have signed up.If you are interested in this option, please contact send an e-mail to secretaria@etsetb.upc.edu with subject “Inscripció al seminari MET 230360 DLCV”, attaching single page motivation letter stating your previous experience with deep learning, as well as a one page CV, both in PDF. Registration is limited to local industry members and students registered in an institutionfrom the European Higher Education Area.


Piazza will be used for class discussion and communication, instead of the regular UPC Atenea platform. Piazza is highly catered to getting you help fast and efficiently from classmates, TAs and instructors. Rather than emailing questions to the teaching staff, I encourage you to post your questions on Piazza. If you have any problems or feedback for the developers, email team@piazza.com.

Find us at the class page.


Previous editions

Related courses


GitHub Education



Universitat Politècnica de Catalunya