| CODE | ARI5118 | ||||||||||||
| TITLE | Deep Learning for Computer Vision | ||||||||||||
| UM LEVEL | 05 - Postgraduate Modular Diploma or Degree Course | ||||||||||||
| MQF LEVEL | 7 | ||||||||||||
| ECTS CREDITS | 5 | ||||||||||||
| DEPARTMENT | Artificial Intelligence | ||||||||||||
| DESCRIPTION | Since 2012, deep learning has been revolutionising the fields of computer vision and artificial intelligence. Deep feedforward networks, supported with vast datasets and computing power can now surpass human capabilities in performing object detection and even facial recognition. Deep learning has proved useful in a variety of disciplines and this study-unit focuses on the state of the art developments in computer vision. The core of this unit focuses on variations of convolutional and recurrent neural networks as applied to vision problems. This study-unit provides this theory with a sound connection to the practical aspect of implementing deep feedforward network by making use of the latest languages, libraries and frameworks. Study-unit Aims: - Introduce deep learning techniques in the context of artificial vision; - Enable students to critically analyse different deep learning approaches; - Enable students to choose the right deep learning approach for a given application; - Provide background about the different layers and architecture of deep feedforward networks; - Expose students to different datasets; - Equip students with skills to be able to implement deep learning approaches in their dissertations and/or other study units. Learning Outcomes: 1. Knowledge & Understanding By the end of the study-unit the student will be able to: - Explain the motivation behind the use of deep feedforward networks in artificial vision; - Demonstrate the importance of using the right data for the right network; - Explain how the network architecture can be modified to achieve better results and be applied to different problems; - Describe how a convolutional neural (CNN) network works; - Differentiate between different variations of CNNs; - Describe how a recurrent neural network (RNN) works; - Differentiate between different variations of RNNs. 2. Skills By the end of the study-unit the student will be able to: - Implement a variety of deep networks for vision using available libraries and frameworks; - Apply deep feedforward networks to artificial vision problem; - Discuss and critically analyse different deep learning architectures; - Decide which deep network architecture is most suitable for a given scenario; - Employ different datasets to train and test deep feedforward networks; - Critically analyse practical methodologies and applications of deep learning in artificial vision scenarios. Main Text/s and any supplementary readings: - Ian Goodfellow and Yoshua Bengio and Aaron Courville. Deep Learning. MIT Press, 2016. Available online: http://www.deeplearningbook.org/ - Michael Nielsen. Deep Learning 2017. available online: http://neuralnetworksanddeeplearning.com Additional Reading - Josh Patterson and Adam Gibson. Deep Learning - a practitioner's approach, O'Reilly, 2017. |
||||||||||||
| RULES/CONDITIONS | Before TAKING THIS UNIT YOU MUST TAKE ICS5110 | ||||||||||||
| STUDY-UNIT TYPE | Lecture and Project | ||||||||||||
| METHOD OF ASSESSMENT |
|
||||||||||||
| LECTURER/S | Dylan Seychell |
||||||||||||
|
The University makes every effort to ensure that the published Courses Plans, Programmes of Study and Study-Unit information are complete and up-to-date at the time of publication. The University reserves the right to make changes in case errors are detected after publication.
The availability of optional units may be subject to timetabling constraints. Units not attracting a sufficient number of registrations may be withdrawn without notice. It should be noted that all the information in the description above applies to study-units available during the academic year 2024/5. It may be subject to change in subsequent years. |
|||||||||||||