IMAGE AND VIDEO UNDERSTANDING

The course aims at introducing the student to the principles, the algorithms and the main applications in the field of image and video understanding.

1. Knowledge and understanding
1.1. acquire the main models and algorithms of image and video understanding

2. Ability to apply knowledge and understanding
2.1. acquire the ability to apply the studied models to real problems
2.2. acquire the ability to critically assess the performance and the behavior of a model applied to a concrete problem

3. Judgement
3.1. ability to understand which characteristics of the various models of artificial intelligence are best suited to a given problem
3.2. ability to critically evaluate the theoretical characteristics of the proposed models

The student is expected to be familiar with the basic concepts of calculus, linear algebra and statistics. Knowledge of Python language, together with PyTorch, are recommended.

Neural Network Models Models:
- Artificial Neural Networks (training, tricks, optimizers)
- Convolutional Neural Networks
- Transformer Architectures
- Graph Neural Networks

Image Analysis:
- Classification
- Segmentation
- Object Detection

Video Understanding:
- Video Classification
- Video Object Segmentation
- Object Tracking

Human-Centered Computer Vision:
- Person detection
- Face detection
- Pose Estimation
- Person Re-Identification
- Trajectory Forecasting
- Action Recognition
- Group Detection

Explainability Techniques
- Image data
- Graph data

Generative AI:
- Auto Encoders & Variational Autoencoders
- GANS
- Diffusion Models
- Flow Matching

Multimodal Deep Learning
- Alignment techniques
- Image and text
- Open vocabulary object detection

Learning paradigms:
- Self-supervised learning
- Reinforcement learning

Foundation models

Advanced Topics (tentative):
- Vision for Robotics
- Active Learning
- Anomaly Detection
- Neural Implicit Representation
- Scene Understanding

- R. Szeliski, Computer Vision: Algorithms and Applications. Springer.

- D. Forsyth and J. Ponce. Computer Vision: A modern Approach. Pearson.

- I. Goodfellow, Y. Bengio and A. Courville. Deep Learning. MIT Press

The exam consists of an oral test (70% of the final mark) together with a discussion of a project (30% of the final mark) agreed upon with the teacher.

oral

The lecturer has a duty to ensure that the rules regarding the authenticity and originality of exam tests and papers are respected. Therefore, if there is suspicion of irregular conduct, an additional assessment may be conducted, which could differ from the original exam description.

A. Scores in the 18-22 range will be awarded in the presence of:
- Sufficient knowledge and ability to structure the project;
- Limited ability to justify implementation choices;
- Sufficient communication skills, especially in relation to the use of course-specific language.

B. Scores in the 23-26 range will be awarded in the presence of:
- Fair knowledge and ability to structure the project;
- Fair ability to collect and/or interpret data, proposing effective implementation solutions;
- Fair communication skills, especially in relation to the use of course-specific language.

C. Scores in the 27-30 range will be awarded in the presence of:
- Good or excellent knowledge and ability to structure the project;
- Good or excellent ability to collect and/or interpret data, proposing innovative implementation solutions;
- Fully appropriate communication skills, especially in relation to the use of course-specific language.

D. Lode will be awarded in the presence of excellent knowledge and applied understanding of the program, judgment skills, and communication abilities.

Powerpoint presentations and chalk talk.

To favor an "active" appraoch to the study of the topics covered in the classes, students will be asked to develop a simple project which will be discussed during the oral examination.