User:Nrp/Weekly updates/20080719: Difference between revisions

From OLPC
Jump to navigation Jump to search
(New page: == Saturday == * Fixed incorrect behavior in an existing pygame function, mask.get_bounding_rects(), and also doubled its performance. == Sunday == * Wrote a function to list cameras avail...)
 
No edit summary
Line 4: Line 4:
* Wrote a function to list cameras available on the computer.
* Wrote a function to list cameras available on the computer.
* Played with integrating pymunk and the camera module. I wrote a script that has gravity go in the direction of a real life object the user holds up. A fun little demo, but it barely scratches the surface of computer vision and physics integration. The next step is integrating it with [[Physics]] or [[Elements]] on the XO.
* Played with integrating pymunk and the camera module. I wrote a script that has gravity go in the direction of a real life object the user holds up. A fun little demo, but it barely scratches the surface of computer vision and physics integration. The next step is integrating it with [[Physics]] or [[Elements]] on the XO.
== Monday ==
* Attempting to package an .rpm
* Played with threading
== Tuesday ==
* Started writing an example script to test the possibility of pixel perfect collision detection between real life and on screen objects.
== Wednesday ==
* Finished a few versions of the script. There are definite opportunities here, though it certainly needs a lot of work to make it more than just a nifty tech demo. There is, as usual, also the problem that it runs at an unplayable speed above 320x240 on the XO. Is there perhaps a way to hook into xv to do hardware scaling for display? Something to look into.

Revision as of 00:11, 17 July 2008

Saturday

  • Fixed incorrect behavior in an existing pygame function, mask.get_bounding_rects(), and also doubled its performance.

Sunday

  • Wrote a function to list cameras available on the computer.
  • Played with integrating pymunk and the camera module. I wrote a script that has gravity go in the direction of a real life object the user holds up. A fun little demo, but it barely scratches the surface of computer vision and physics integration. The next step is integrating it with Physics or Elements on the XO.

Monday

  • Attempting to package an .rpm
  • Played with threading

Tuesday

  • Started writing an example script to test the possibility of pixel perfect collision detection between real life and on screen objects.

Wednesday

  • Finished a few versions of the script. There are definite opportunities here, though it certainly needs a lot of work to make it more than just a nifty tech demo. There is, as usual, also the problem that it runs at an unplayable speed above 320x240 on the XO. Is there perhaps a way to hook into xv to do hardware scaling for display? Something to look into.