I teach Fundamentals of Computer Programming for the Bachelor Degree in Mechanical and Industrial Engineering. In this course we learn how to design basic algorithms and how to code such algorithms in the C language. Basic data structures (list, trees), sorting and searching algorithms are covered.
Lectures are typically a mixture of teaching, active learning and live coding. Students are encouraged to complete voluntary homework during teaching period.
All content is constantly updated on moodle: enroll (no password required)
Introduction to Generative Adversarial Networks with M. Bertini – a.y. 2018/2019
In this course we will provide both technical and theoretical insights on how to build and train generative adversarial networks. We will cover basic unconditioned generation, conditional generation and image2image problems such as style transfer, super resolution and compression artifact removal.
Introduction to Deep Learning in Keras with M. Bertini and T. Uricchio – a.y. 2017/2018
In this hands-on course we start from simple MLP models, we than introduce Convolutional Neural Networks and finally we give a quick introduction to generativ adversarial networks. The whole course is based on colab notebook based on Keras.
Visual Recognition: From Handcrafted Features to End-to-End Learning with L. Ballan – a.y. 2016/2017
In this course we first review local handcrafted features and quantization based pipelines. Finally we show how to learn visual features directly from data using convolutional neural networks.
My research is focussed on the use of deep learning for computer vision. If you are interested in topics such as: image generation/enhancement using GANs, object and action recognition, vehicle path prediction contact me email to discuss available projects. You may have a look at the following master and bachelor theses final projects to get an idea of the topics.
Take a look at my posts on my blog for summaries of my recent research results.
- Andrea Amelio Ravelli (ongoing)
- Federico Becattini, graduated in 2017, now PostDoc at MICC.
- Leonardo Galteri, graduated in 2017, now PostDoc at MICC.
- Andrea Ciamarra (ongoing)
- Lorenzo Berlincioni (ongoing)
- Francesco Marchetti, “Vehicle Trajectory Prediction”, April 2019
- Luca Cultrera (September 2018)
- Silvia Palozzi, July 2015
- Andrea Zerbini, July 2015
- Federico Becattini, October 2014, MTAP 2017
- Leonardo Galteri, October 2014 IEEE TIP 2017, ICPR 2018
- Andrea Ciolini, October 2014, “Efficient Hough Forest Object Detection For Low-Power Devices”, ICME-WS 2015
- Enrico Bondi, “Crowd counting and analysis system via depth sensor”, Apr. 2014, AVSS2014
- Claudio Baecchi&Francesco Turchini , “Fisher Feature Fusion Forests for visual object recognition”, Jun. 2013, ICPR 2014
- Leonardo Galteri, “Real-time low level feature extraction on a surveillance camera”, Dec. 2011.
- Lorenzo Usai, “Hand pose recognition with Kinect™“, Mar. 2012, ICPR 2012
- Vincenzo Varano,”Action Recognition from 3D cameras”, Apr. 2013, CVPR-WS 2013
Past Courses and Tutorials
In this tutorial we will build a bag of words pipeline from scratch. Attendees will get an overview of this popular method including several practical and implementation details. We will try to code some simple steps of the pipeline in order to gain insight on the method. Code, data and pre-computed features are available on the website. A full working system will also be provided with all the solutions to the exercises proposed during the tutorial. The provided data is a subset (4 and 15 categories) of Caltech-101 dataset; the pipeline is compatible with the full Caltech-101 and Caltech-256 directory structure with no further modification.
Digital Circuits – Polo Universitario Aretino del Politecnico di Milano
- Interactive digital circuits simulation software: logisim-evolution
- Open source VHDL compiler and simulator: ghdl
- Slides on VHDL
- Example circuits using VHDL components:
Lectures for Multimedia Databases (DBMM)
- Intro to obj. categorization 8/11/2012 [ pdf ]
- SVM classification 8/11/2012[ zip(pdf+libsvm) ]
- Event detection 25/11/2011 [ pdf ]
- Object Categorization 18/11/2009 [ppt] [ pdf ]
- Space-time features 21,25/11/2011 [ pdf ]
- PLSA 21/11/2011 [ pdf ]
- Expectation Maximization 21/11/2011 [ pdf ] [ code ]
- Object recognition with SURF and SIFT* examples [ code ]
- DBMM09 Contest: logo recognition with SIFT [ code & dataset ]
- Introduction to OpenCV [ pdf ] [ code ]
- Human Action Recognition and Event Detection 19/11/2009 [ ppt ] [ pdf ]
- Simple application to test HSV color histogram [ code ]
*derived from Rob Hess’ SIFT.