Hands-on Part I – Computer Vision basics, theory, and tools
Instructor: Benoit Seguin (from Image and Visual Representation Lab – | École Polytechnique Fédérale de Lausanne)
An introduction of basic notions about the challenges of computer vision. A feeling of the simple, low-level operations necessary for the next stage.
Tools:
Python
Basic image operations: scikit-image
Face-object identification + identification: dlib
Deep Learning: Keras
What is CV?
How to gain high-level understanding from digital images or videos.
It tries to resolve tasks that humans can do (Wikipedia)
Human Vision System (HVS) versus Digital Image Processing (what the computer sees)
Practice:
– Jupyter system (an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and explanatory text);
– perform basic image operations;
– Play with different convolutions to develop intuition.
Hands-on Part II – Deep Learning and its application
- Info: https://hackmd.io/s/Hkx3hAxPW
- Visualisation links: http://setosa.io/ev/image-kernels/
- http://scs.ryerson.ca/~aharley/vis/conv/
- Imagenet Challenge: accuracy increased from 28,2& errors to 3% in 6 years. More on Imagenet in this talk of Fei-fei Li https://qz.com/1034972/the-data-that-changed-the-direction-of-ai-research-and-possibly-the-world/ Slides of her talk http://image-net.org/challenges/talks_2017/imagenet_ilsvrc2017_v1.0.pdf