Module COMP52715: Deep Learning for Computer Vision and Robotics

Department: Computer Science

COMP52715: Deep Learning for Computer Vision and Robotics

Prerequisites

• None

Corequisites

• COMP52815 Robotics - Planning and Motion; COMP52615 Computer Vision; PHYS51915 Introduction to Machine Learning and Statistics; PHYS52015 Introduction to Scientific and High Performance Computing

Excluded Combination of Modules

• None

Aims

• Develop knowledge of key concepts, approaches and algorithms for the use of recent advances in deep machine learning applied to tasks within the context of computer vision and robotics.
• Develop critical understanding and appreciation of current theoretical and empirical research in the use of deep machine learning approaches within the context of computer vision and robotics and its application within industry.

Content

• Themes will be chosen from contemporary areas of deep machine learning applied to tasks within the context of computer vision and robotics including the following:
• scene reconstruction and understanding from multiple images or video;
• scene reconstruction and understanding from active sensing;
• Simultaneous Localisation and Mapping (SLAM) from varying sensor inputs;
• visual odometry from varying sensor inputs;
• robotic guidance and control;
• contemporary and emerging research and applications.

Learning Outcomes

• By the end of the module students should have:
• developed a critical understanding of the contemporary deep machine learning topics presented, how these are applicable to relevant industrial problems and have future potential for emerging needs in both a research and industrial setting;
• developed an advanced knowledge of the principles and practice of analysing relevant robotics and computer vision deep machine learning based algorithms for problem suitability;
• developed a good understanding of managing the trade-off between task performance and processing requirements within the context of robotics and computer vision systems;
• explored the most recent advancements in the relevant academic literature and developed a critical understanding of their implications for current industry practice.

• By the end of the module, students should have developed highly specialised and advanced technical, professional and academic skills that enable them to:
• formulate and solve problems that involve the use of contemporary deep machine learning approaches within the context of robotics and computer vision tasks using a range of algorithmic approaches;
• develop software solutions that make use of contemporary deep machine learning approaches to address both industrial and research application tasks within the context of robotics and computer vision.

• Written communication;
• Planning, organising and time management;
• Problem solving and analysis;
• Using initiative
• Adaptability
• Numeracy
• Computer literacy

Modes of Teaching, Learning and Assessment and how these contribute to the learning outcomes of the module

• A combination of lectures, seminars, and guided reading will contribute to achieving the aims and learning outcomes of this module.
• The summative written assignment will test students' knowledge and critical understanding of the material covered in the module, their analytical and problem-solving skills.

Teaching Methods and Learning Hours

Summative Assessment

Formative Assessment:

Feedback on coursework.

■ Attendance at all activities marked with this symbol will be monitored. Students who fail to attend these activities, or to complete the summative or formative assessment specified above, will be subject to the procedures defined in the University's General Regulation V, and may be required to leave the University