Sparx

From Computer Laboratory Group Design Projects
Revision as of 17:44, 13 October 2018 by afb21 (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

Have concluded that it is too far to come to Cambridge for 3 meetings.

Last contacts with Glenn Woodcock <glenn@sparx.co.uk> and Tanya Morton <Tanya.Morton@sparx.co.uk>


2018

OpenFace is an open source facial behaviour analysis toolkit, that can monitor a webcam to detect emotional state via "action units”, such as smiling or raised eyebrows. Imagine a game that could adapt itself based on the emotions of the player. For example, it could display helpful hints when it sees the player is confused, or make it harder when the player looks bored. Your task is to implement such a game for the classroom. The choice of genre is up to you, for example a puzzle game with hints and different levels, or a platform game with procedurally generated content.

Feedback:

I wondered if there might be an opportunity to update this, following recent interest in the iPhone X FaceID system, which has been described as being used to select or animate emoji’s? Rather than focusing on when the player looks bored (which is technically difficult - there are few facial signs other than an open mouth yawn that are both detectable and associated with boredom), they could create a game which uses a range of emotions.


Last year's discussion with Glenn proposed:

I Love Physics

OpenFace is an open source facial behavior analysis toolkit that can monitor a webcam to detect emotional state via "action units" such as smiling or raised eyebrows. Imagine an online tutorial system that watched your face, and could provide assistance as soon as it sees that you are stuck. This would be an alternative to children's toys such as Osmo (which adapts an iPad camera for reality-based teaching games - www.playosmo.com). Your task is to implement a physics teaching game inspired by Osmo's Newton, but where objects bounce off features of the user's face, and it waits for a happy smile when you have understood the equations!

2017 proposal

The client will be Tanya Morton and Will Bolam ((will.bolam@sparx.co.uk).

Another idea from me:

I Love Physics

OpenFace is an open source facial behavior analysis toolkit that can monitor a webcam to detect emotional state via "action units" such as smiling or raised eyebrows. Imagine an online tutorial system that watched your face, and could provide assistance as soon as it sees that you are stuck. This would be an alternative to children's toys such as Osmo (which adapts an iPad camera for reality-based teaching games - www.playosmo.com). Your task is to implement a physics teaching game inspired by Osmo's Newton, but where objects bounce off features of the user's face, and it waits for a happy smile when you have understood the equations!

Based on two ideas from Sparx:

1. Crowd-sourcing Human Feedback

As we move towards a world in which the use of technology is ubiquitous within education, it’s crucial to retain the personalised and insightful feedback that only a human can provide. Your task is to build a system that uses crowd sourcing techniques to integrate human feedback into an otherwise automated system. The system should allow any question to be asked - think video, audio or image as well as text based questions. Similarly the system should support answers in any format the user would like - written or drawn answers could be input using an Osmo, or a smart phone could capture video, audio or images. The answer should be uploaded to the cloud ready for marking by another human anywhere in the world. The marker should be able to overlay their feedback and comments within the answer provided, and then return the answer and feedback to the original user. We’ll provide an Osmo as inspiration.

2. Emotional Learning

Does emotional state and heart rate impact learning? Recent advances in wearable biosensors (think fitbit) have made this a much easier question to answer. Your task is to design a system to explore this question, using biorhythm monitoring to track emotional state whilst your subjects complete various tasks. Exactly how you test people is up to you, it could be anything from a simple maths quiz to a complex game with different physical and mental tasks ­ of course the more engaging it is the more data you will get! At the end you’ll need to analyse the data and spot correlations in order to form a conclusion and answer the question. We’ll provide you with a small budget for suitable wearable tech.

Earlier idea from me:

Data Science for Kids

Although public decisions in the UK often depend on statistical analysis, the public are not encouraged to get involved. Future voters could learn to participate in democratic processes involving data that will have personal consequences for them, applying evidence to local policy debates such as speed limits and road design. Your task is to create an easily accessible tool for kids to become democratic data scientists, helping their families analyse public experiments such as Cambridge traffic surveys, going beyond the usual comparison of averages to apply modern data science techniques such as time series, clustering, analysis of variance, linear regression etc.

Original suggestion:

Ready for Secondary Maths?

The government has identified that some students are not secondary ready by the time they leave primary school. Mathematical ability is a key factor in this assessment. Sparx works closely with Year 7 students, some of who have a low level of mathematical skill. Sparx is keen to explore how to support these students in their final year of primary school and better prepare them for entering secondary school.

Low ability students struggle with basic topics such as: number bonds, addition, subtraction and multiplication. Your challenge is to design an engaging and personalised software tool to help teach and re-enforce these skills. The software should include data visualisation capabilities to enable the teacher to assess student progress.

2015 project

Flash Mob Learning


Idea 2:

Running a classroom of tablets in a school over their wireless system can be pretty volatile.

In order to teach a dynamic lesson with 30 students using individual iPads, can we create a wireless, router-less system? Can we create a network of tablets without needing a router? How many can we connect at once? What are the distance limitations?

The implications of this for Sparx could be extensive, because if you have a group of cheap mobile devices, but no other infrastructure – as is the case in many developing countries- this solution could be the difference between accessing education and not.

Idea 1:

“You might not know this, but over 40% of 16 year olds fail their GCSE maths, in Britain, each year. KPMG has calculated that this costs the UK, £2.4 billion every year.

Sparx has been working for two years with 15 and 16 year olds to find out how we learn, and how to improve performance. We have been developing a platform which can teach them in the way that they respond to best.

We know that for many of the students we have worked with, they don’t have basic number sense or their times tables locked down as “learnt’ yet. Can you create a game which would appeal to the 11 – 16 year old market, which would teach times tables, whilst helping the teacher to understand what is and isn’t secure in their learning? Can you develop a game that improves their learning and can give feedback to them and the teacher and can extend each student, in a personalised and adaptive way?"