openworm / movement_cloud

Movement Analysis on the cloud
http://movement.openworm.org/
Other
5 stars 4 forks source link

Summarize repo relationships in a diagram #94

Open MichaelCurrie opened 7 years ago

MichaelCurrie commented 7 years ago

It turns out in February 2016 I had started working on this already:

https://www.lucidchart.com/documents/edit/544ac7aa-763a-49b8-a860-15734084f845#

@aexbrown you mentioned we should get this updated for the paper. I will work on this. Everyone else please feel free to make edits as needed.

MichaelCurrie commented 7 years ago

One issue I have with these diagrams is what relationships to depict. Is it data flow, data relationships, or something else?

image

ver228 commented 7 years ago

Hello Michael, I am not currently using the Tracking_Hypo.

Currently it is like:

experiments -> tierpsy <-> OWAT 
                  v
                 WCON

I guess in the future it could be something like:

 experiments -> (any tracker) ->  WCON -> OWAT    
MichaelCurrie commented 7 years ago

@ver288 does this mean tierpsy-tracker contains all the computer vision code to turn the videos into skeleton data? Is tracking_hypo obsolete since the code is in tierpsy?

In the future could we provide video processing to other labs also?

experiments -> (any tracker) -> tierpsy video processing -> WCON -> OWAT

ver228 commented 7 years ago

No, we haven't really integrated tracking_hypo work. I think they were some issues left.

MichaelCurrie commented 7 years ago

This workflow could be shared in an explanatory page on the website as a landing page for anyone interested in Worm Tracker 2.0, tierpsy tracker, WCON, or OWAT.

It would be nice to illustrate each of the one-liners needed to start these steps from the command line, as well as what to do from the website.