Introducing lftextures
December 4, 2009 at 18:29
filed under Light Field
Tagged light_field, opengl, software
| Download: | Windows (3M) |
| Linux 32 64 (3M) | |
| Mac (3M) | |
| Datasets: | Path (2M), Landscape (3M) |
Here we have available some binaries of our simple light field viewer, lftextures. When Dan and I first tried to capture a light field, trying to manipulate the data was a real pain. How were we to know if our data was any good if we didn’t have a simple test to show we had captured the scene? We decided to implement the easiest operation, digital refocusing. Here’s the product:
The controls are simple, use the mouse wheel or Home/End keys to scroll through the scene; the application refocuses as you scroll.
Scroll up = move to background, Scroll down = move to foreground.
- Mouse wheel, or Home/End keys: Scroll through scene
- Click and drag mouse: Rotate the viewing plane
- Arrow keys: Move the viewing plane
- Page Up/ Page Down: Zoom in/out from the viewing plane
- R: Reset the viewing pane
- P: Print the screen (files are dropped in the directory lftextures is in, i.e. the working directory)
- O: Turn off/on the on-screen display (The text in the main window)
- X: Speed up/down the rate of scrolling
- L: Brighten/Darken the scene
- M: Switch render modes
And finally,a few macros (for making movies)
- I: Print screen and Scroll up once
- U: Print screen and Scroll down once
- Y: Print screen and rotate away
- T: Print screen and rotate towards
Once you’re satisfied scrolling through the scene, try hitting “M” until the on-screen display says “M: 2, 3D” then click and drag the scene back and forth:
DrAltaica
on December 19, 2009 at 21:26
What algorithm are you using to refocus the image? I don’t get why objects appear to move right as the focus moves further away.
matti
on December 19, 2009 at 23:06
To tell the truth, Dan and I are still mystified that the algorithm I wrote works so well. It nudges each image to the right, but each successive image gets pushed further than the one before it.
for(i=0; i < NUM_PIC; i++)
{
glTranslatef( offset*i, 0, 0);
drawImage( i );
}
I could have moved the screen as the image moved, but I didn’t.
If you hit ‘m’ once, you get to ’slide’ mode, which doesn’t really work, but was supposed to be the same idea. The slide mode fixes the center image and then moves the other images around it, the images that come before it go one way and the images that come after it go the other direction.
DrAltaica
on December 20, 2009 at 08:56
glTranslatef is opengl function?
so you are just shifting each image over by X * the camera number them averaging them together?
matti
on December 20, 2009 at 13:09
Yes, and yes. That is exactly how the rendering algorithm works.
Surprising, huh? It was the very first thing I tried, to debug the display to see if all the images were on screen, moved, and blended with each other. Imagine my surprise when the debug routine actually refocused the image!
» Welcome! > FUTUREPICTURE
on December 21, 2009 at 13:19
[...] GIFs look really bad. Here are some links to higher-resolution images if you don’t want to install our software and play with the included images (or your [...]
Jeremy
on December 21, 2009 at 15:17
Linux 64 bit is 404
Jeremy
on December 21, 2009 at 15:45
My friend and I were discussing how the out-of-focus portions of the photo contain multiply copies of an object. We’ve decided to dub it “meta-bokeh”
matti
on December 21, 2009 at 15:50
Fixed the 64-bit link, and the landscape dataset.
DrAltaica
on December 22, 2009 at 00:19
@matti:
I redid you refocusing techniqu in PhotoShop
http://img193.imageshack.us/i/photographtest.jpg/
then I took a Tomograph of the lightfield(@ “Plane of focus” of the refocused image I made)
http://img44.imageshack.us/i/tomographytest.jpg/
Not much of a difference between them
http://img684.imageshack.us/i/diff.jpg/
If you guys could take a picture of a flat object. Like a book or pizza box just far enough away so that it’s compleatly in view of all the cameras and perpendicular to the otical axis of your camara(So it’s the same shape on the photo as it is in real life). It would help differ
I think it’s the ratio of the Entrance Pupil’s size to the distance to the Plane of Focus.
I wonder how you are going to compensate for the fact that you are not using pinhole camera?
DrAltaica
on December 22, 2009 at 02:03
I just realized I messed up on the refocusing.
Here is the fixed Refocusing.
http://img85.imageshack.us/i/photofixed.jpg/
and the fixed Differance
http://img705.imageshack.us/i/fixeddiff.jpg/
You can see the differance on the tire mark on the path.