Frontier: Difference between revisions

From Computer Laboratory Group Design Projects
Jump to navigationJump to search
No edit summary
Line 2: Line 2:


Also with Eddy Aston in 2017 <eashton@frontier.co.uk>
Also with Eddy Aston in 2017 <eashton@frontier.co.uk>
==Project proposals for 2018:==
-Simulation LOD-ing
Level-of-detail (LOD) optimisations have a long history in rendering. 3D
meshes and textures are simplified as objects move away from the camera,
avoiding wasteful computation where a user can no longer distinguish finer
details. There is increasing use of LOD-ing in other areas such as animation
and AI routines, allowong us to simulate large, active crowds with high
frequency details in the foreground. Can we use similar techniques to
simulate other systems which are otherwise too complex or large to be fully
computed? Perhaps a global weather simulation interacting with person-sized
physics objects, or a densely populated world with interactions ranging from
person-to-person up to empire-to-empire. The challenge is in finding a
generalised representation of the simulation at small scales which has a
stable, consistent approximation at larger scales, allowing a user to zoom
to any point and see deterministic results. Can you include user input to
alter the simulation, and simulate that change at different scales and over
time? You may wish to do this on constrained hardware such as a Raspberry
Pi, where an otherwise simple simulation may be too complex and so the gains
of LOD-ing are necessary.
-Opposing Views
Many people increasingly exist in an ideological bubble, getting all their
news from a handful of sources that they trust and shunning everything else.
This leads to polarised, insular viewpoints, stifling debate and increasing
divisiveness. Your task is to make a web browser plugin which encourages
users to read about a story from multiple angles. When reading an article,
it can suggest an alternative, opposing source for the same topic. It needs
to be able to categorise stories so that it can recognise the same topic in
different publications, categorise sources so that it can find an
interesting alternative stance (ideally not just how left- or right-leaning
it is, but also things like the background of the author, the size and age
of the publication, the tone of the articles), and also present this to the
user in a way which will circumvent their defensive instincts and encourage
an open mind.
-VR AI Ping Pong Trainer
There have been lots of ideas involving uses of AI and VR in recent years,
and we would like to offer you the chance to work on a project that brings
these two areas together. We would like you to create a VR application
displaying a simple view of a Table Tennis table, bats, and ball, and an
environment which may track the ball and administer a game of Ping-Pong.
However, we would like the opponent to be an AI which you are attempting to
coach. It may be useful to have the option to run fast automatic training
also to refine your bots, but the ultimate aim is to be able to provide some
human training, and to be able to play them directly in VR. It may also be
interesting to have the option to pitch two separately trained bots against
each other and see how they behave. Rather than arcade games where there are
a range of buttons to press with different functions, ping pong offers only
the position, rotation and motion of a bat as it's means of control,
providing an interesting input mechanism to work with.
Notes:
With the Ping Pong Trainer, we realise the motion control may require some
consideration. We are open to any options you think suitable, and also did
wonder whether it may be possible to use a common Nintendo Wiimote with
supporting software such as GlovePie or similar (after seeing articles such
as
https://www.howtogeek.com/howto/35470/how-to-use-your-wii-remote-as-a-gyrosc
opic-mouse/). It may not be suitable, or there may be better options, but it
was just something we wanted to throw in the pot, just in case.


==Project proposals for 2017:==
==Project proposals for 2017:==
Line 28: Line 99:
   
   
   
   
.. and slightly more high level
and slightly more high level
   
   
Plan -> VR
Plan -> VR

Revision as of 06:31, 5 October 2017

The client is Matt Johnson (mjohnson@frontier.co.uk)

Also with Eddy Aston in 2017 <eashton@frontier.co.uk>

Project proposals for 2018:

-Simulation LOD-ing

Level-of-detail (LOD) optimisations have a long history in rendering. 3D meshes and textures are simplified as objects move away from the camera, avoiding wasteful computation where a user can no longer distinguish finer details. There is increasing use of LOD-ing in other areas such as animation and AI routines, allowong us to simulate large, active crowds with high frequency details in the foreground. Can we use similar techniques to simulate other systems which are otherwise too complex or large to be fully computed? Perhaps a global weather simulation interacting with person-sized physics objects, or a densely populated world with interactions ranging from person-to-person up to empire-to-empire. The challenge is in finding a generalised representation of the simulation at small scales which has a stable, consistent approximation at larger scales, allowing a user to zoom to any point and see deterministic results. Can you include user input to alter the simulation, and simulate that change at different scales and over time? You may wish to do this on constrained hardware such as a Raspberry Pi, where an otherwise simple simulation may be too complex and so the gains of LOD-ing are necessary.


-Opposing Views

Many people increasingly exist in an ideological bubble, getting all their news from a handful of sources that they trust and shunning everything else. This leads to polarised, insular viewpoints, stifling debate and increasing divisiveness. Your task is to make a web browser plugin which encourages users to read about a story from multiple angles. When reading an article, it can suggest an alternative, opposing source for the same topic. It needs to be able to categorise stories so that it can recognise the same topic in different publications, categorise sources so that it can find an interesting alternative stance (ideally not just how left- or right-leaning it is, but also things like the background of the author, the size and age of the publication, the tone of the articles), and also present this to the user in a way which will circumvent their defensive instincts and encourage an open mind.


-VR AI Ping Pong Trainer

There have been lots of ideas involving uses of AI and VR in recent years, and we would like to offer you the chance to work on a project that brings these two areas together. We would like you to create a VR application displaying a simple view of a Table Tennis table, bats, and ball, and an environment which may track the ball and administer a game of Ping-Pong. However, we would like the opponent to be an AI which you are attempting to coach. It may be useful to have the option to run fast automatic training also to refine your bots, but the ultimate aim is to be able to provide some human training, and to be able to play them directly in VR. It may also be interesting to have the option to pitch two separately trained bots against each other and see how they behave. Rather than arcade games where there are a range of buttons to press with different functions, ping pong offers only the position, rotation and motion of a bat as it's means of control, providing an interesting input mechanism to work with.


Notes:

With the Ping Pong Trainer, we realise the motion control may require some consideration. We are open to any options you think suitable, and also did wonder whether it may be possible to use a common Nintendo Wiimote with supporting software such as GlovePie or similar (after seeing articles such as https://www.howtogeek.com/howto/35470/how-to-use-your-wii-remote-as-a-gyrosc opic-mouse/). It may not be suitable, or there may be better options, but it was just something we wanted to throw in the pot, just in case.

Project proposals for 2017:

Creature Dash

In some video games over the years, and most notably of recent in No Man's Sky, attempts have been made to make procedural alien life forms, to give a sense of diversity and variety, yet avoiding the production difficulties of manually creating and animating such a vast range of potential life forms. A common thread in these products solutions has been to come up with a malleable connected graph that can represent a creature out of a small and finite number of given parameterised nodes.

The challenge is to build a platform which provides a number of parameterised 'primitive parts' which may be connected together by users in building block fashion to create automotive creatures with some kind of walk, or propellant movement cycle. The users may also use other provided simple building blocks to construct 'obstacle' courses to commit their creatures to, to explore their success in navigating them, and to see which creature finishes first. Innovations involving sensors, planning, navigation and movement are welcome, but must extend the modular parameterised system, rather than being specific to an individual creature. The only rule is there must be no user interaction with individual creatures during its progress on the course.

My version:

VR Algorave DJ

Why can't a DJ be more like an orchestra conductor, remixing instrumental sections and adding new expressive content, instead of simply selecting from a library of prerecorded tracks? In the future this will be possible, using a gesture controllers in virtual reality. We can provide a Myo gesture control armband. Your task is to create a VR space in which the DJs of the future will edit and configure algorave-style music synthesis programs. You can use APIs for the Sonic Pi music language from Cambridge's Sam Aaron as a back end to produce professional-standard musical results.


VR DJ

Virtual Reality is becoming an increasingly popular medium for computing applications, particularly in entertainment. There is currently a lot of undiscovered potential in VR, especially related to it's use in creativity and manipulation, rather than just in terms of 'experience'. This project aims to explore the more tactile and immediate possible uses for VR, to use a headset and application to create live, responsive audio/visual entertainment, through the use of virtual controls presented to the user, to allow them to mix audio, video and potentially other media presented to an audience in real-time.


VR Minigame BuildSpace

VR has become a serious contender in the field of video games and entertainment in the last few years. There is still limited applications for it in the this area which allow users to build and share their own games in a common virtual environment. Your task in this project will be to build on the backs of some existing technologies to develop such a space, allowing users to build operational games out of primitives, assets and rules, and potentially even collaborating to do so.


and slightly more high level

Plan -> VR

This project is about interpreting 2d building plans given in a commonly accepted format, and using this to produce a decorated and lit 3d environment from the plan, allowing walls and faces to be set with various materials, and so rendered as such from a palette of textures and models. The environments should be navigable via a convincing VR interface, potentially with the ability to manipulate and decorate them, from within

Projects in 2016:

Equity Exchange

Projects in 2015:

Projects 2014:

Projects in 2013:

Contact: Ben Nicholson bnicholson@frontier.co.uk