• OK, it's on.
  • Please note that many, many Email Addresses used for spam, are not accepted at registration. Select a respectable Free email.
  • Done now. Domine miserere nobis.

A programming problem

Cogitant

Fiducial Observer
Local time
Today 12:28 PM
Joined
Jun 8, 2017
Messages
157
-->
Location
Invading your reality
For those who enjoy graphical modelling, it has been asked of me to come up with some ideas for my brother's latest software project.
What he needs is to design a detailed, dynamic program which can accurately track motion and translate precise visual data into an ipad/phone format (quadcore or so).

Not allowed to surrender details regarding what he's up to (his work = Fight Club) but what I can tell you is that it's possibly for some kind of a scanner.

"Panning and zoom are super straight forward
also dynamic mesh modification"

https://en.wikipedia.org/wiki/Point_cloud => https://en.wikipedia.org/wiki/Octree

Just thought some of you might be interested.
Any relevant ideas and links are greatly appreciated.

Thanks.
 

Cognisant

Prolific Member
Local time
Today 12:28 AM
Joined
Dec 12, 2009
Messages
10,593
-->
I don't understand, you're talking about motion capture but your links are about 3D modeling, does he want to do 3D scanning of an object in motion? For what purpose?
 

Cogitant

Fiducial Observer
Local time
Today 12:28 PM
Joined
Jun 8, 2017
Messages
157
-->
Location
Invading your reality
Those are the links he provided me with.

I've just now bombarded him with some questions on Steam in order to clarify. His response:

profound: it does involve getting a single mesh of someone using a camera that moves
profound: handheld IR camera.

It sounds a little abstract, I know, but he only throws scraps of information in my direction (he expects me to understand/work out concepts from a few words). I have a clue what he might be looking at. I'll start writing up.
 

baccheion

Active Member
Local time
Today 7:28 AM
Joined
May 2, 2016
Messages
277
-->
I'm fairly certain vision tracking libraries/APIs already exist.
 

gps

INTP 5w4 Iconoclast
Local time
Today 7:28 AM
Joined
Mar 16, 2010
Messages
200
-->
Location
Upstate NY, USA, Earth
Regardless of the particular app, wouldn't it be interesting and potentially useful IF `code' could evaluated via this site ... perhaps as a front end which outsources the actual evaluation?

I've been using org-mode for years and have been spoiled by it's ability to embed various source codes within an org-mode file which is many ways is similar to a post to a vBullitin group.

Imagine having a `run' or `execute' button available to press while the cursor is hovering over a bbcode `code' block.
This would allow those with `ideas' to submit code which could be empirically tested by others in the group as a more high-level alternative to presenting mere textual descriptions.
 

Ex-User (14663)

Prolific Member
Local time
Today 12:28 PM
Joined
Jun 7, 2017
Messages
2,939
-->
Regardless of the particular app, wouldn't it be interesting and potentially useful IF `code' could evaluated via this site ... perhaps as a front end which outsources the actual evaluation?

I've been using org-mode for years and have been spoiled by it's ability to embed various source codes within an org-mode file which is many ways is similar to a post to a vBullitin group.

Imagine having a `run' or `execute' button available to press while the cursor is hovering over a bbcode `code' block.
This would allow those with `ideas' to submit code which could be empirically tested by others in the group as a more high-level alternative to presenting mere textual descriptions.
Why not just compile/run it locally?
 

washti

yo vengo para lo mío
Local time
Today 1:28 PM
Joined
Sep 11, 2016
Messages
862
-->
because when you put it on orbit it tastes the best
 

gps

INTP 5w4 Iconoclast
Local time
Today 7:28 AM
Joined
Mar 16, 2010
Messages
200
-->
Location
Upstate NY, USA, Earth
Why not just compile/run it locally?
Once again ...

This would allow those with `ideas' to submit code which could be empirically tested by others in the group as a more high-level alternative to presenting mere textual descriptions.
For all or most would care it ultimately could be executed by local CPU cycles the way most `try-' implementations are via client-side javascript running in one's browser.
(EG tryScheme, tryAPL, papertLogo, etc)

For a related suchness check out reproducible research.

From an ease-of-use perspective, not everybody participating here has compilers, interpreters and such installed on whatever gadget they use to access this site; so if they could just tap on a key to join in the fun and games they could participate on equal footing with those who would take the time to copy the source code, paste it into a file, and work with it locally on a machine capable of compiling, interpreting, or whatever.
Many will AVOID the aforementioned HASSLES, but would find tapping a `run' or `try it' button a mere speed bump to be taken in stride rather than a hurdle to be leaped over.
Think about it; how many will use a search engine to perform research when a mere text string is provided as contrasted with clicking on a link?
The `run' or `try it' functionality would allow folks all the convenience of clicking on a link.

If we INTPs who have programmed and enjoy it want to facilitate our INTP brethren who don't or haven't-yet TRYING something which involves programming then how better to grease the skids and remove possibly-unnecessary obstacles?
 
Top Bottom