4th Summer School
on Deep Learning for Vision
UPC ETSETB TelecomBCN (June 25 - July 3, 2019)
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.
This Summer School will include a special day of oral presentations of local researchers with a high international impact.This session will be located in the Aula Master of A3 building in Campus Nord UPC (directions). These talks will be open to the general public for free, but registration will be required via the following link.
Tuesday, June 25 @ D5-010
- 15:00 D1L1 Welcome (XG)
- 15:25 Project: Snakes Fine-grained Classification (XG, KM, FR, OM & BO)
- 16:25 Break
- 16:35 D1L2 Neural Architectures for Still Images (XG) [Slides]
- 17:00 D1L3 Object Detection (AG)
- 17:25 Break
- 17:35 D1L4 Face Detection & Recognition (RM)
- 18:00 D1L5 Neural Architectures for Videos (XG)
- 18:25 D1L6 Object Tracking (AG)
- 19:00 End of the day
Wednesday, June 26 @ D5-010
- 15:00 D2L1 3D Analysis (JR)
- 15:25 D2L2 Image segmentation (AP)
- 15:50 D2L3 Instance Segmentation (AP)
- 16:15 Break
- 16:35 D2L4 Retrieval (KM)
- 17:00 D2L5 Visual Generation I (AP)
- 17:25 D2L6 Visual Generation II (AP)
- 18:00 Project: Snakes Fine-grained Classification (KM, FR, OM & BO)
- 19:00 End of the day
Thursday, June 27 @ D5-010
- 15:00 D3L1 3D Reconstruction I (ER)
- 15:25 D3L2 3D Reconstruction II (ER)
- 15:50 Project: Snakes Fine-grained Classification (KM, LL, FR, OM & BO)
- 16:50 Break
- 17:15 D3L3 Similarity Learning (LL)
- 17:45 D3L4 Bayesian Deep Learning (LL)
- 18:20 D3L5 Neural Network Visualization (LL)
- 19:00 End of the day
Friday, June 28 @ Aula Master A3 (Open Day)
- 15:00 Petia Radeva (UB-CVC), “Food Image Analysis - a perfect test to understand what Transfer learning can do”
- 15:25 Adrià Arbués (UPF), “Single-Camera Basketball Tracker through Pose and Semantic Feature Fusion” (ICAIS 2019)
- 15:50 Kevin McGuinness (DCU-Insight), “Unsupervised label noise modeling and loss correction” (ICML 2019)
- 16:15 Carles Ventura (UOC), “RVOS: End-to-End Recurrent Network for Video Object Segmentation” (CVPR 2019)
- 16:45 Break
- 17:00 Pau Riba (UAB-CVC), “Doodle to Search: Practical Zero-Shot Sketch-based Image Retrieval” (CVPR 2019 - oral)
- 17:25 Laura Leal-Taixé (TUM), “Understanding the Limitations of CNN-based Absolute Camera Pose Regression” (CVPR 2019)
- 17:50 Albert Pumarola (UPC-IRI), “Deep Learning for Human-Centrinc Applications in AR/VR.”
- 18:15 Amaia Salvador (UPC-IDEAI), “Inverse Cooking: Recipe Generation from Food Images” (CVPR 2019)
- 18:15 End of the day
Monday, July 1 @ D5-010
- 15:00 D4L1 Medical Imaging (ES)
- 15:25 D4L2 Saliency (KM)
- 15:50 D4L3 Noisy labels (KM)
- 16:15 Project: Snakes Fine-grained Classification (KM, FR, OM & BO)
- 17:50 Break
- 18:00 D4L4 Vision and Language (XG)
- 19:00 End of the day
Wednesday, July 3 @ D5-010
- 15:00 Exam (KM)
- 16:30 Student project expos (KM)
- 18:50 Closing & homework (KM)
The students developed a hands on challenge proposed by Dr. Rafael Ruiz de Castañeda and the Institute of Global Health in Geneve (Switzerland) in order to use Deep Learning to detect snake species to prevent snake bites, which cause of 100.000 human deaths each year. This project was selected as on of the most interesting in the recent CVPR 2019 Workshop on Computer Vision for Globall Challenges.
- Teaching language: English
- Class Schedule: 3pm-7pm (you will need 4 extra hours a day for homework during the week course)
- Capacity: 20 students
- Location: Campus Nord UPC, Module D5, Room 010
- Course code: 230376 DLV
- Class Dates: June 25 - July 3, 2019
- ECTS credits: 3.0 (corresponds to full-time dedication during the week course)
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 Winter School on Introduction to Deep Learning at the end of January , or signing up for the full master course of Deep Learning for Artificial Intelligence taught during Autumn.
If you can prove previous training in deep learning (eg. DLAI, IDL, or ML&DL CFIS):
Master students at ETSETB: Registration is available from the ETSETB academic office. There is an extraordinary registration around the end of May. 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”. You will need to first register through this form before 10 June 2019. After the course sign up for the credits at the ETSETB academic office.
CFIS students: You must have completed 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 or researchers at UPC & Master in Computer Vision: You might audit the course, with no official certification. If interested, fill in this form before 10 June 2019. You may be granted a seat, if there are any available after ETSETB, CFIS & FIB students have signed up. UPDATE (20th June 2019): The course is full now. If you have not received an e-mail from Piazza confirming your seat, it means that we could not allocate your request.
Other students currently enroled in another academic program: You can request attending to the seminar by sending your CV and motivation letter to the ETSETB academic office, clearly showing your previous knowledge in deep learning and their software frameworks. If accepted, you will need to cover the full cost of the course (429,24 euros, which correspond to 3 ECTS during academic course 2018/2019). You must alreaday be legally eligible to attend to this course, we will not release any acceptance later for visas.
Other profiles (eg. industry members): This course is not addressed to this audience. If you are interested in this topic, we suggest you consider the postgraduate program in Artificial Intelligence with Deep Learning from UPC School. This postgraduate extends the contents of this Summer School and is also supported by the UPC TelecomBCN School. You can nevertheless attend to the Open Day session on Friday, which is public.
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 email@example.com.
Find us at the class page.
- Deep Learning for Computer Vision UPC TelecomBCN.  
- Deep Learning for Speech and Language UPC TelecomBCN.  
Deep Learning for Video. Master in Computer Vision Barcelona. 
- Deep Learning for Multimedia. Insight Dublin City University 2017.  
- Introduction to Deep Learning. UPC TelecomBCN 2018.
- Deep Learning for Artificial Intelligence. UPC TelecomBCN 2017.
- Amaia Salvador and Santiago Pascual. “Hands on Keras and TensorFlow”. Persontyle 2017.
- Fei-Fei Li, Andrej Karpathy, Justin Johnson, “CS231n: Convolutional Neural Networks for Visual Recognition”. Stanford University, Spring 2019.
- Sanja Fidler, Machine Learning in Computer Vision. University of Toronto, Winter 2019.
- Dhruv Batra, “ECE 6504: Deep learning for perception”. Virginia Tech, Fall 2015.