Visual Intelligence (VI) is another word for modern Computer Vision (CV), i.e. CV based on DL (ML/AI)

Teacher: Frank Lindseth

Methods covered

• Fundamentals (neurons, layers, networks and back-propagation etc.)
• Image Classification (CNNs and backbones)
• Object Detection (R-CNN++, YOLO++, SSD, etc.) and Tracking
• Segmentation (Mask R-CNN, U-Net, etc.) and Pose Estimation
• Vision Transformers and Diffusers
• Depth estimation, super resolution and style transfer.
• Deep Generative Models (DGM) and CV
• Deep Reinforcement Learning (DRL) and CV
• From image to 3D content / mesh and NeRFs
• etc.

Application examples

• Autonomous Perception
• Medical Image Computing / Analysis
• etc.

Classical and State-of-the-Art papers.

2023:

We will have our startup meeting Thursday Sep 7. at 15:00 next week on Teams (everybody that has signed up will get a teams-invite in due time).

The plan is to have three "seminars" during the semester (we will use Thursdays for that as well, the exact dates will be given later).

The background of the people who have signed up is probably a bit broad, it's an advantage that we are somewhat on the same page, if you know that you are a bit rusty on things like calculus/partial derivatives/lin.algebra, Python/NumPy/PyTorch/Notebooks and DL/NNs please try to read up a bit, a lot of resources are availble online and is a search away, as for the DL/NN part a gentle intro can be:

• Neural Networks (NN, 3Blue1Brown)
• Neural Networks and Deep Learning (Nielsen)

More info and how to sign up for non-IDI students can be found here:

Info, Modules, TDT4506 (Computer Science), IT3020 (Informatics), Registration (other study programmes)

2022:

Update: 

• Things seem to have stabilized a bit now and most students have decided which modules to attend, but if YOU are not in this category and want TDT17 after Sep 6 at 15:00 drop me an email (frankl@ntnu.no).
• Those who have already signed up will receive an email later today. I have received quite some messages that the suggested day is not ideal but we mainly need to do this on Wed., you will decide if it will be in the start or end of the day though, the poll will be out soon. 
• There is also students having challenges with attending this week for various reasons but there will always be someone that can not join in
• All in all, and since this update was supposed to be sent out on Sunday, and not late today, I have decided to have the first meeting Wed Sep 14 next week (either 08:15 or 14:15). Until then you will get some more info to get you going.

---
Welcome to the 2022-edition of this module.
This module will probably not start until Wed Sep 7th (updated info will be given here).
Experience show that the students need some time in order to decide on what modules to attend.
And my calendar is not completely set yet.
The intention is to have a mix of digital and physical activity.
The first meeting will be digitally on Teams.
Additional information will be provided at the first meeting.
Registration to the module for students external to IDI should be done using this link:
https://www.idi.ntnu.no/education/mine_teorimoduler.php

The background of the attending students can vary quite significantly and we need to find a common ground. Basic skills in calculus, linear algebra and programming are needed. A good indication of knowhow and intuition expected before the first meeting can be found in the two links below. If you already have this that's good and you are all set, if this is not the case and you want to attend the module you should try to gain this before the module formally start (this is also why we have a late start).
Neural Networks and Deep Learning (Nielsen)
Building micrograd (Karpathy)

Looking forward to get started.