The area of computer vision is ever-expanding from medical applications to robotic systems to smartphone technology and beyond. This is due to increased processing power, improved quality and reduced cost of vision hardware and the advancement of AI and image processing techniques. This module provides an introduction to computer vision techniques and applications in real-world contexts.
Visual Data
An introduction to the type of data inputs used in Computer Vision: digital images, video, 3D data, etc. This will include a look at capture technologies, colour spaces, pixels, voxels and videos and involve working with visual data in a coding environment, accessing pixels, annotating, handling user input, etc.
Analysis
Understanding and interpreting visual data. This will include spatial and temporal analysis, colour and visual analysis, histograms, optical flow, 3D transformations, etc.
Transformation
Techniques for processing and modifying visual data, including convolution, histogram manipulations, mathematical operations, etc. This will also include simple manipulation methods such as resizing, cropping, rotating, etc.
Features
Working with data features at both low level (edges, corners, flow vectors, etc) and high level (objects, motion, etc). This will include feature detectors and feature descriptors as well as matching, tracking and clustering of features.
Image Retrieval
Using the vector-space model and techniques such as image-vocabulary creation and indexing to build efficient algorithms for image retrieval
Image Classification
Employing machine learning libraries to build image classification models such as K-nearest neighbours, Bayes classifiers and support vector machines and convolutional neural networks
Computer Vision Applications
Implementation of the techniques learnt to real-world applications such as object detection, motion tracking, image enhancement, restoration and measurement.
A combination of in-class tasks and group project work
Material is presented online and in-person
Assessment is done online as software designs, project presentations and in-class tests, assessed using detailed bespoke rubrics.
Assessment includes peer assessment and self-assessment.
| Module Content & Assessment | |
|---|---|
| Assessment Breakdown | % |
| Other Assessment(s) | 100 |