Processing Forum

IndexSizeError: DOM Exception 1 (Processing.js, Aruco.js, getUserMedia)

[1 Reply]
- 16-Apr-2013 02:19 AM
- Forum: Processing with Other Languages
I am attempting to use the Aruco.js augmented reality library with Processing.js. I already have it's most simple implementation working, where I am using getUserMedia to use the webcam through the browser and applying it to the canvas using context.drawImage(). I then use context.getImageData() to grab the color information and feed it to Aruco via detector.detect(). This gives me Marker objects, which provide polygon information which I then read in my processing.js sketch to draw. The result looks like this (the red polygon borders, green marker, and blue number are all rendered using processing methods):

Of course, the next step is to get the rotation information for each AR marker in order to do some cool 3D stuff. This is where I ran into my problem. In order to use rotateX() and rotateY() I need to setup my processing sketch using the OPENGL context. When I switch to OPENGL, I can no longer use the context.drawImage() method that was applying the video frame to the canvas. So I ended up making one canvas (set to display: none) which acts as a container for my frames in order to get the color data, and another canvas for my Processing.js WebGL content.

When I launch the page, I can see my processing sketch with a little spinning cube (in order to insure the draw loop is still working) and the prompt at the top of the browser asking to allow the camera. As soon as I hit allow, my animation stops and I get this error in the browser console:
1. Uncaught Error: IndexSizeError: DOM Exception 1 processing.js:4076
So, I start commenting out code and I get to the point where I discover the line that is actually causing the issue is this one:
1. imageData = context.getImageData(0, 0, canvas.width, canvas.height);
However, I can't find any information regarding this DOM exception or issues with using multiple canvas objects with Processing.js. I am testing this in Chrome 26.0.1410.64m on Windows 7 and Chrome 26.0.1410.65 on OSX 10.7.5 using Processing.js 1.4.1. The error is consistent across platforms. Here is the complete code:
processing_3dar.html
2. <!DOCTYPE html>
3. <html lang="en">
4. <head>
5. <title>processingjs augmented reality test</title>
6. <meta charset="utf-8">
7. <link rel="stylesheet" type="text/css" href="_css/basic.css" />
8. </head>
10. <body>
11. <script src="_js/cv.js"></script>
12. <script src="_js/aruco.js"></script>
13. <script src="_js/posit1.js"></script>
14. <script src="_js/svd.js"></script>
15. <script src="_js/processing.js"></script>
16. <script src="_js/aruco3dp5.js"></script>
18. <video id="video" autoplay="true" style="display:none;"></video>
19. <canvas id="canvas" data-processing-sources="_pde/processing_aruco3d.pde" width="960" height="720"></canvas>
20. <canvas id="imagecanvas" width="960" height="720" style="display:none;"></canvas>
21. </body>
23. </html>
aruco3dp5.js:
2. var video; // video DOM element
3. var canvas; // canvas DOM element
4. var context; // canvas javascript context
5. var imageData; // webcam image data
6. var detector; // AR detector
7. var markers; // array of marker objects for any given frame
8. var posit; // posit object to analyze corners for rotation vector
9. var markerSize = 18;
11. /*
12. * Init function.
13. */
14. function onLoad(){
15. // establish DOM elements and canvas context
16. video = document.getElementById("video");
17. canvas = document.getElementById("imagecanvas");
18. context = canvas.getContext("2d");
20. // get canvas dimensions
21. canvas.width = parseInt(canvas.style.width);
22. canvas.height = parseInt(canvas.style.height);
24. // get access to camera
25. navigator.getUserMedia = navigator.getUserMedia || navigator.webkitGetUserMedia;
26. if (navigator.getUserMedia){
27. navigator.getUserMedia({video: true}, successCallback, errorCallback);
29. function successCallback(stream){
30. if (window.webkitURL) {
31. video.src = window.webkitURL.createObjectURL(stream);
32. } else {
33. video.src = stream;
34. }
35. }
37. function errorCallback(error){
38. // do nothing, I guess.
39. }
41. // create the AR detector
42. detector = new AR.Detector();
44. // create the Posit object
45. posit = new POS.Posit(markerSize, canvas.width);
46. }
48. }
50. /*
51. * Grab the current frame from the video.
52. */
53. function snapshot(){
54. context.drawImage(video, 0, 0, canvas.width, canvas.height);
55. // FIXME: line below: "Uncaught Error: IndexSizeError: DOM Exception 1"
56. imageData = context.getImageData(0, 0, canvas.width, canvas.height);
57. //markers = detector.detect(imageData);
58. }
60. window.onload = onLoad;
processing_aruco3d.pde (note that drawOutline and drawPose are never called for purposes of testing):
2. void setup(){
3. size(960, 720, OPENGL);
4. frameRate(30);
5. rectMode(CENTER);
6. strokeWeight(3);
7. }
9. void draw(){
10. if (video.readyState === video.HAVE_ENOUGH_DATA){
11. snapshot();
12. //drawOutline();
13. }
15. background(255);
16. translate(width/2, height/2);
17. rotateY(0.5);
18. rotateX(frameCount * 0.01);
19. box(100);
20. }
22. void drawOutline(){
23. for(Object m : markers){
24. // draw outline
25. noFill();
26. stroke(255,0,0);
27. beginShape();
28. for(int i=0; i<m.corners.length; i++){
29. vertex(m.corners[i].x, m.corners[i].y);
30. }
31. endShape(CLOSE);
33. // draw primary corner
34. stroke(0,255,0);
35. rect(m.corners[0].x, m.corners[0].y, 4, 4);
37. // draw ID number
38. fill(0,150,255);
39. text(m.id, m.corners[0].x, m.corners[0].y);
41. // draw pose
42. //drawPose(m.corners);
43. }
44. }
46. void drawPose(Object[] corners){
47. // translate corners to center justify marker face
48. for(Object corner : corners){
49. corner.x = corner.x - (width / 2);
50. corner.y = (height / 2) - corner.y;
51. }
52. Object pose = posit.pose(corners);
54. // get rotation and translation values
55. Object rot = pose.bestRotation;
56. Object trans = pose.bestTranslation;
58. // convert rotation values to radians (I think)
59. float yaw = -Math.atan2(rot[0][2], rot[2][2]);
60. float pitch = -Math.asin(-rot[1][2]);
61. float roll = Math.atan2(rot[1][0], rot[1][1]);
63. // draw a cube
64. pushMatrix();
65. translate(trans.x, trans.y, trans.z);
66. rotateX(pitch);
67. rotateY(yaw);
68. rotateZ(roll);
69. box(markerSize);
70. popMatrix();
71. }

Processing core video:217): GStreamer-CRITICAL **

[3 Replies]
- 04-Dec-2012 03:56 AM
- Forum: Core Library Questions
Never encountered this one before. After resolving my applications issues in this thread I decided to move the application bundle over to an installation computer hooked up to a large display in the hallway of my work. I had to swap out the video library native libs for the 32 bit ones, and then change the camera device being selected, and then I got this string of really strange errors:

** (Processing core video:217): WARNING **: ColorConverter: size 62400 is not a multiple of unit size 60000

2012-12-04 12:41:11.523 JavaApplicationStub[217:903] invalid drawable

(Processing core video:217): GStreamer-CRITICAL **:

Trying to dispose element nat, but it is in PAUSED instead of the NULL state.

You need to explicitly set elements to the NULL state before

dropping the final reference, to allow them to clean up.

This problem may also be caused by a refcounting bug in the

application or some element.

(Processing core video:217): GStreamer-CRITICAL **:

Trying to dispose element Video Capture, but it is in PAUSED instead of the NULL state.

You need to explicitly set elements to the NULL state before

dropping the final reference, to allow them to clean up.

This problem may also be caused by a refcounting bug in the

application or some element.

I can't seem to find anything existing in the forum about this. I am running it on a Mac Mini 1.66ghz Intel Core Duo with 1gb of ram running OS X 10.6.8 and the Processing 2.0b6 core and video libraries.

Eclipse / Processing 2.0 / Video / Mac OSX Application Bundle / UnsatisfiedLinkError: Unable to load library 'gstreamer-0.10'

[11 Replies]
- 26-Nov-2012 05:47 AM
- Forum: Integration and Hardware
Once again approaching the subject of the gstreamer native libraries, I have been through a number of forum posts (such as this and this) and have found nothing that seems to work for my situation. Here is my scenario:

I have been developing this application in Eclipse 4.2.0 on OSX 10.7.4 with Processing 2.0b6, in which it runs fine. When I attempt to create an application bundle, it does not work. My application uses the Capture class to get video, run some optical flow analysis on it to create a flow field, and then use that to push around some particle objects. Here is what it looks like when it runs.

The structure of my Eclipse project is as follows:

InteractiveScreen/

InteractiveScreen/src

InteractiveScreen/data

InteractiveScreen/lib

Inside the "src" folder I have my package and the three classes, InteractiveScreen.java, ColorSpectrum.java, and Particle.java. Inside my "data" folder there are a couple pngs for textures and a main.props file for loading in variables such as screen size, particle attributes, etc. Inside my "lib" folder I have:

core.jar

gluegen-rt-natives-macosx-universal.jar

gluegen-rt.jar

gstreamer-java.jar

jna.jar

jogl-all-natives-macosx-universal.jar

jogl-all.jar

macosx32/ (contains all of the necessary native libraries for video)

video.jar

All of which are added to the build path, except for gluegen-rt-natives-macosx-universal.jar, jogl-all-natives-macosx-universal.jar, and macosx32/. In my build path, I have gluegen and jogl referencing the "lib" directory for their native library location, and I have gstreamer referencing the "lib/macosx" directory for it's native library location.

Once again, all of this setup runs fine in Eclipse. Now, I export an application bundle via the " Export > Other > Mac OS X application bundle" menu. The directory structure for that is as follows:

InteractiveScreen.app

InteractiveScreen.app/Contents/

InteractiveScreen.app/Contents/Info.plist

InteractiveScreen.app/Contents/MacOS/

InteractiveScreen.app/Contents/MacOS/JavaApplicationStub

InteractiveScreen.app/Contents/PkgInfo

InteractiveScreen.app/Contents/Resources/

InteractiveScreen.app/Contents/Resources/data/

InteractiveScreen.app/Contents/Resources/data/cloud.png

InteractiveScreen.app/Contents/Resources/data/cloud2.png

InteractiveScreen.app/Contents/Resources/data/main.props

InteractiveScreen.app/Contents/Resources/data/sphere.png

InteractiveScreen.app/Contents/Resources/InteractiveScreen.icns

InteractiveScreen.app/Contents/Resources/Java/

InteractiveScreen.app/Contents/Resources/Java/core.jar

InteractiveScreen.app/Contents/Resources/Java/gluegen-rt-natives-macosx-universal.jar

InteractiveScreen.app/Contents/Resources/Java/gluegen-rt.jar

InteractiveScreen.app/Contents/Resources/Java/gstreamer-java.jar

InteractiveScreen.app/Contents/Resources/Java/jar_0.jar

InteractiveScreen.app/Contents/Resources/Java/jar_1.jar

InteractiveScreen.app/Contents/Resources/Java/jna.jar

InteractiveScreen.app/Contents/Resources/Java/jogl-all-natives-macosx-universal.jar

InteractiveScreen.app/Contents/Resources/Java/jogl-all.jar

InteractiveScreen.app/Contents/Resources/Java/macosx32/

InteractiveScreen.app/Contents/Resources/Java/video.jar

Inside the InteractiveScreen.app/Contents/Resources/Java/macosx32 directory is the massive list of native .dylib files and plugins directory which I will refrain from posting here.

My Info.plist file reads as such:
1. <?xml version="1.0" encoding="UTF-8" standalone="no"?>
2. <!DOCTYPE plist PUBLIC "-//Apple Computer//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
3. <plist version="1.0">
4. <dict>
5. <key>CFBundleExecutable</key>
6. <string>JavaApplicationStub</string>
7. <key>CFBundleGetInfoString</key>
8. <string>InteractiveScreen1.0, Copyright datadreamer 2012. All rights reserved.</string>
9. <key>CFBundleInfoDictionaryVersion</key>
10. <string>6.0</string>
11. <key>CFBundleIconFile</key>
12. <string>InteractiveScreen.icns</string>
13. <key>CFBundleIdentifier</key>
14. <string>InteractiveScreen</string>
15. <key>CFBundleName</key>
16. <string>InteractiveScreen</string>
17. <key>CFBundlePackageType</key>
18. <string>APPL</string>
19. <key>CFBundleSignature</key>
20. <string>????</string>
21. <key>CFBundleVersion</key>
22. <string>1.0.1</string>
23. <key>Java</key>
24. <dict>
25. <key>WorkingDirectory</key>
26. <string>$APP_PACKAGE/Contents/Resources/</string>
27. <key>JVMVersion</key>
28. <string>1.6*</string>
29. <key>MainClass</key>
30. <string>it.fabrica.interactivescreen.InteractiveScreen</string>
31. <key>Arguments</key>
32. <array/>
33. <key>VMOptions</key>
34. <string>-d32 -Xmx1024m -Djava.library.path=macosx32</string>
35. <key>ClassPath</key>
36. <string>$JAVAROOT/jar_0.jar:$JAVAROOT/core.jar:$JAVAROOT/gluegen-rt.jar:$JAVAROOT/jogl-all.jar:$JAVAROOT/video.jar:$JAVAROOT/gstreamer-java.jar:$JAVAROOT/jna.jar:$JAVAROOT/jar_1.jar:$JAVAROOT/core.jar:$JAVAROOT/macosx32</string>
37. </dict>
38. </dict>
39. </plist>
Note the WorkingDirectory entry where I set the "$APP_PACKAGE/Contents/Resources/" folder as the working directory, and also the VMOptions entry, where I am forcing the JVM to run in 32 bit mode to accommodate the video library, set the heapspace to allow for 1024mb of memory usage, and added the macosx32 folder to the library path. I am also adding the macosx32 folder to the ClassPath entry, just in case. Unfortunately, neither seem to be making a difference with finding the native libraries for gstreamer. No matter where I place those files, I still get this error:

Exception in thread "AWT-EventQueue-0" java.lang.UnsatisfiedLinkError: Unable to load library 'gstreamer-0.10': dlopen(libgstreamer-0.10.dylib, 9): image not found

at com.sun.jna.NativeLibrary.loadLibrary(NativeLibrary.java:163)

I would really like to figure out what the issue is and how to resolve it so that I can put together a more comprehensive solution for creating application bundles with Processing/Video in the future.

GLGraphicsOffScreen with Transparent Background

[1 Reply]
- 25-Feb-2012 04:12 AM
- Forum: Contributed Library Questions
I remember in the past creating an off screen object to draw on and filling/erasing the background with all transparent pixels at the beginning of every draw(), instead of simply redrawing background(color). I can't recall if it was a PImage or PGraphics, but ultimately I would like to accomplish this with a GLGraphicsOffScreen object.

I realize I can probably just iterate over the pixels and force each one to have an alpha of 0, but I remember using a single method to accomplish this (though I don't remember the type of object I was using nor can I seem to find it any of my projects).

TLDR: What's the best way to make a GLGraphicsOffScreen object always have a transparent background?

Two GLGraphics Windows + Two Video Cards? Hardware requirement/optimization question.

[0 Replies]
- 15-Feb-2012 08:51 PM
- Forum: Integration and Hardware
I am working on a project that will be using GLGraphics as the renderer, and utilizes two windows for output. One window is a small control panel window, and the second uses a GLTextureWindow to display everything rendered to a GLGraphicsOffScreen object. The GLTextureWindow will span 8 displays (anywhere between 1280x720 each to 1920x1080 each), giving us a total resolution of the GLGraphicsOffScreen canvas from 10240x720 to 15360x1080.

We plan to get a Radeon Eyefinity 6 to give us 6 outputs, and then use a Matrox TripleHead2Go on the 6th output to give us a total of 8. We also want to have the control panel window on a 9th display that will be installed with the computer for the exhibition, and allows us some control over parameters.

My concern is regarding the sharing of an OpenGL application across GPUs. Our options are (1) buy a secondary cheap card that will serve as the 9th output for the control panel, or (2) use a DoubleHead2Go on another one of the outputs to keep the rendering all on a single video card.

When I have a single window with an OpenGL context in it, and I run it on my workstation with 2 video cards, and I make it span two displays (each on their own card) I end up with very choppy results (which is to be expected). However, I have never created two windows before, one being the primary GLGraphics rendering and the other being the GLTextureWindow. Ultimately, my question is this: will rendering the two different windows on two different video cards result in the choppiness due to GPU sharing, or will they behave in a manner where one window uses one video card and the other window uses the other video card?

glBlendFunc cross platform issue

[8 Replies]
- 20-Sep-2011 09:47 PM
- Forum: Contributed Library Questions
I am encountering a strange cross-platform issue while using glBlendFunc. I developed this on Windows XP but it will ultimately be running in installation on a MacBook pro. My contact person at the installation site has been testing the software and has noticed a very strange blending problem that doesn't occur on my computer. I have included a comparison screenshot of the issue, with the left side captured on Windows XP and the right side captured on the Mac.

As you can see, the blending works well in Windows, but on the Mac the textures appear to have their transparent areas saturated black. I am using gl.glBlendFunc(GL.GL_SRC_ALPHA, GL.GL_DST_ALPHA); to accomplish this type of blend. I am using GLGraphics as the renderer. The texture is being rendered as a Texture object like so:
2. // draw indicator
3. gl.glPushMatrix();
4. gl.glTranslatef(x, y, 0);
5. gl.glBlendFunc(GL.GL_SRC_ALPHA, GL.GL_DST_ALPHA);
6. gl.glColor4f(r, g, b, a);
7. texture.bind();
8. texture.enable();
10. gl.glBegin(GL.GL_QUADS);
11. gl.glTexCoord2f(0, 0);
12. gl.glVertex2f(-diameter/2, -diameter/2);
13. gl.glTexCoord2f(1, 0);
14. gl.glVertex2f(diameter/2, -diameter/2);
15. gl.glTexCoord2f(1, 1);
16. gl.glVertex2f(diameter/2, diameter/2);
17. gl.glTexCoord2f(0, 1);
18. gl.glVertex2f(-diameter/2, diameter/2);
19. gl.glEnd();
21. texture.disable();
22. gl.glPopMatrix();
My machine is running Windows XP 64-bit with a GeForce 9800GT. The mac is running Snow leopard 10.6.8, has an I ntel core i7 2.2ghz, and AMD Radeon HD 6750m and Intel HD Graphics 3000. I unfortunately don't have a mac of my own to test different blend options on to see what will work, so I am relying on hopefully somebody having encountered a similar cross-platform issue before.

Super Large 9600 x 1080 GLGraphicsOffScreen object

[9 Replies]
- 06-Jul-2011 06:38 PM
- Forum: Contributed Library Questions
Here is a rundown of my current project:

5 large 1080p displays hooked up to a Radeon 6990. We will attempt to use an Eyefinity 5x1 configuration (if we can ever figure out how to properly set it up) with a bit of a twist. Since we want the displays to be side by side in landscape mode and run along a wall in a VERY wide setup (around 20+ feet), and since Eyefinity 5x1 mode only supports portrait mode (5400 x 1920), I intend to draw to a raster off screen using GLGraphics which will be 9600 x 1080, and then split that into 5 pieces, rotate them 90 degrees, and then place them side by side. The effect is similar to this:

This is a 960 x 108 (1/10th the width and height, ie: 1/100th the final intended size) drawn to a GLGraphicsOffScreen object. When I actually apply it to the GLGraphics context being displayed, which is only 540 x 192, it looks like this:

When all of the displays are rotated 90 degrees into landscape mode, it should all look perfect.

Still with me? Great.

This works as expected at this resolution. However, I am increasing to production size in order to build a jar and send it out to be tested on the production machine with the displays attached. The workstation arrived today featuring an i7 950, 6 gigs of ram, the Radeon 6990, and Windows 7 pro 64-bit. It's a beast. However, I won't be back to work with the hardware myself for another 2 weeks, so I am stuck developing precursor test jars on my laptop (a 1.5 year old thinkpad... not terrible, but no beast).

In preparation for this, I increased my size() parameter to 5400 x 1920, and my off screen size to 9600 x 1080. When I ran my application, I received this error:

Frame buffer is incomplete (GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT)

OpenGL error 1280 at top endDraw(): invalid enumerant

Frame buffer is incomplete (GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT)

OpenGL error 1282 at render_triangles in: invalid operation

OpenGL error 1286 at non-binding 6: invalid framebuffer operation

Frame buffer is incomplete (GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT)

Frame buffer is incomplete (GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT)

Frame buffer is incomplete (GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT)

Frame buffer is incomplete (GL_FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT_EXT)

... and then it just repeats the last line ad infinitum while the application runs.

Having searched the forum for this type of error I don't find a whole lot of results. So I figure one of two things: (1) my relatively puny graphics acceleration chip in my laptop just can't handle the awesomeness, or (2) this is some kind of constraint in either GLGraphics or OpenGL itself... I really don't know.

ANY insight into what these errors mean, or limitation in the software, or even the hardware I'm using, would be MUCH appreciated. Do you think I can build the jar and send it off and it may work on the awesome 6990 hardware, and it's just a limitation of my laptop? Or am I boned thinking that I can execute this entirely?

Included are two snippets of code from my project thus far; NevadaMain.java and NevadaApplet.java. NevadaMain is simply the frame to house my PApplet instance, and NevadaApplet extends PApplet (sorry PDE users, but I only develop in Eclipse now). I have also included the props file which is being read to determine my resolution settings.

NEVADAMAIN.java:
2. package com.datadreamer.nevada;
4. import java.applet.Applet;
5. import java.awt.event.WindowAdapter;
6. import java.awt.event.WindowEvent;
7. import java.io.File;
8. import java.io.FileInputStream;
9. import java.io.FileNotFoundException;
10. import java.io.IOException;
11. import java.util.Properties;
13. import javax.swing.JFrame;
15. @SuppressWarnings("serial")
16. public class NevadaMain extends JFrame {
18. private Applet app;
20. public NevadaMain() {
21. super("Nevada Discovery Museum");
22. Properties props = loadProperties("data/main.props");
23. int w = Integer.valueOf(props.getProperty("displayWidth"));
24. int h = Integer.valueOf(props.getProperty("displayHeight"));
25. setResizable(false);
26. setBounds(0, 0, w, h);
27. setUndecorated(true);
28. setVisible(true);
29. app = new NevadaApplet(w, h, props);
30. add(app);
32. addWindowListener(new WindowAdapter() {
33. public void windowOpened(WindowEvent e) {
34. app.requestFocus();
35. }
37. public void windowClosing(WindowEvent e) {
38. System.exit(0);
39. }
40. });
42. app.init();
43. }
45. public Properties loadProperties(String file) {
46. Properties newprops = new Properties();
47. try {
48. newprops = new Properties();
49. newprops.load(new FileInputStream(new File(file)));
50. } catch (FileNotFoundException e) {
51. e.printStackTrace();
52. } catch (IOException e) {
53. e.printStackTrace();
54. }
55. return newprops;
56. }
58. public static void main(String[] args) {
59. new NevadaMain();
60. }
62. }
NEVADAAPPLET.java:
2. package com.datadreamer.nevada;
4. import java.util.Properties;
6. import processing.core.PApplet;
7. import processing.core.PImage;
8. import codeanticode.glgraphics.GLConstants;
9. import codeanticode.glgraphics.GLGraphicsOffScreen;
10. import codeanticode.glgraphics.GLTexture;
12. @SuppressWarnings("serial")
13. public class NevadaApplet extends PApplet {
15. private final int w, h;
16. // private final Properties props;
17. private GLGraphicsOffScreen canvas;
18. private final int offscreenWidth;
19. private final int offscreenHeight;
20. private PImage cloudTest;
22. public NevadaApplet(int w, int h, Properties props) {
23. // this.props = props;
24. this.w = w;
25. this.h = h;
26. offscreenWidth = Integer.valueOf(props.getProperty("offscreenWidth"));
27. offscreenHeight = Integer.valueOf(props.getProperty("offscreenHeight"));
28. }
30. public void setup() {
31. size(w, h, GLConstants.GLGRAPHICS);
32. frameRate(60);
33. canvas = new GLGraphicsOffScreen(this, offscreenWidth, offscreenHeight);
34. cloudTest = loadImage("cloudTest8192.png");
35. }
37. public void draw() {
38. background(255, 0, 0);
39. canvas.beginDraw();
40. canvas.image(cloudTest, 0, 0, offscreenWidth, offscreenHeight);
41. canvas.endDraw();
42. GLTexture fullScreen = canvas.getTexture();
44. translate(0, height);
45. rotate(radians(-90));
46. image(fullScreen, 0, 0, offscreenWidth, width / 5);
47. image(fullScreen, -(offscreenWidth / 5), (width / 5), offscreenWidth,
48. width / 5);
49. image(fullScreen, -(offscreenWidth / 5) * 2, (width / 5) * 2,
50. offscreenWidth, width / 5);
51. image(fullScreen, -(offscreenWidth / 5) * 3, (width / 5) * 3,
52. offscreenWidth, width / 5);
53. image(fullScreen, -(offscreenWidth / 5) * 4, (width / 5) * 4,
54. offscreenWidth, width / 5);
55. }
56. }
MAIN.props
2. # on screen display resolution
3. displayWidth = 5400
4. displayHeight = 1920
5. #displayWidth = 540
6. #displayHeight = 192
8. # off screen canvas being drawn to
9. offscreenWidth = 9600
10. offscreenHeight = 1080
11. #offscreenWidth = 960
12. #offscreenHeight = 108
Toggle the values above to render the 1/100th version without issue, or attempt to strangle your hardware/software with the full version. The image "cloudTest8192.png" is unfortunately quite large (8192 x 922, as that's the widest my laptop's graphics card accepts) and I should probably refrain from uploading it, but I just grabbed a super giant photo of clouds from google images in order to test this.

Also, if you figure out a solution, I will buy you a beer.

TextureIO.newTexture() and "No OpenGL context current on this thread"

[1 Reply]
- 29-Jun-2011 04:59 PM
- Forum: Core Library Questions
Here is a description of what I'm trying to accomplish:

I am creating an application which loads in a series of images sequentially on the fly and displays them using OpenGL. The images are loaded in a separate thread, and each time one is loaded the image is passed to a callback method in a listener object in my processing thread. The image is added to a linked list, and continually checks on draw if the new image has been loaded and can be displayed. The end result is a collection of images stacked in z-space, with the older images moving back and the newer ones fading-in in front.

I got this working successfully, however I was running into a memory leak issue similar to the one described here:

https://forum.processing.org/topic/memory-leaks-with-pimage-loadpixels-updatepixels-image-in-opengl

Since I do not yet want to use an unstable release of processing, I attempted to find a workaround by using direct OpenGL calls to display the image as a texture similar to how it is done here (reply #5 specifically):

http://processing.org/discourse/yabb2/YaBB.pl?num=1199399046/0

However, I still want to be able to load the image in a separate thread so that the rest of my graphics being displayed continue to do so without hanging or stuttering. With this in mind, I am loading a BufferedImage in the separate thread, passing it to the callback method, and then attempting to convert it to a Texture object using TextureIO:
1. texture = TextureIO.newTexture(photo, true);
I get this error when I run it:
javax.media.opengl.GLException: No OpenGL context current on this thread

at javax.media.opengl.glu.GLU.getCurrentGL(GLU.java:234)

at com.sun.opengl.util.texture.Texture.<init>(Texture.java:179)

at com.sun.opengl.util.texture.TextureIO.newTexture(TextureIO.java:445)

at com.sun.opengl.util.texture.TextureIO.newTexture(TextureIO.java:539)

at edu.mit.senseable.backtalk.Photo.<init>(Photo.java:69)

at edu.mit.senseable.backtalk.PhotoStack.imageLoaded(PhotoStack.java:145)

at edu.mit.senseable.backtalk.ImageLoader.run(ImageLoader.java:41)

A little more detail about the classes:

PhotoStack exists at the top level of my class extending PApplet.

ImageLoader extends Thread and it's instance exists within PhotoStack.

PhotoStack implements the ImageListener to provide a callback method.

Photo is instantiated within PhotoStack whenever a new BufferedImage has been passed into PhotoStack via the callback.

Photo attempts to create the Texture object within it's constructor.

I'm at a bit of a loss at this point attempting to set the current OpenGL context. It's my understanding that the object utilizing TextureIO should already be running within the OpenGL context (of which there is only one). Any advice/assistance would be greatly appreciated.

PeasyCam Flickers on Reset

[2 Replies]
- 27-Mar-2011 01:21 AM
- Forum: Contributed Library Questions
I'm using the PeasyCam camera.reset(long duration) method to reset my camera to the original position after a brief period of inactivity, and I'm noticing that upon the reset call I get a weird flicker for one frame of my scene from a completely different perspective. It appears as if it's snapping to the default position from the constructor without applying any of my translations/rotations, before it begins smoothly transitioning to where it should be.

Has anyone else experienced this?

I'm using OpenGL on Mac OS X 10.6.

PeasyCam Conical Movement

[9 Replies]
- 26-Mar-2011 12:24 AM
- Forum: Contributed Library Questions
I am attempting to use the PeasyCam library to accomplish a birds eye view of my subject matter while constantly rotating in a circular movement above it. The camera should maintain a downward view towards the subject while essentially tracking in a circle.

At first I tried simply rotating the camera from looking directly towards my subject plane to the downward angle I desired. I then tried to apply a continual camera.rotateZ(value) in the draw loop, but since the peasycam library uses a type of arcball matrix transformation, it simply causes that current view to be rotated on the new Z axis instead of the world Z axis.

I could use a sin/cos function to make the camera orbit the subject plane by continually updating its position on each frame, as well as update the camera.lookAt() to make sure it is constantly pointed at the same x/y/z position of my subject, but I have my doubts about how the roll of the camera will be handled.

Some people may suggest simply doing a translate to center everything in the world and rotating the subject itself, but I would ideally like to keep all view changes isolated within the camera object.

Has anyone had experience doing this?

Animating BezTube along Bezier Curve

[4 Replies]
- 22-Mar-2011 07:25 AM
- Forum: Programming Questions
I'm developing a project in which I am drawing bezier curves in 3D space. Currently I am animating along these curves using the bezierPoint function to get a number of positions along the curve and then draw vertices at each point. This works great, however now I want to add some varying thickness to the curves. When I change the stroke weight, I notice it varies in quality and style depending on OpenGL implementation. I would like to use actual 3D shapes to draw these curves.

I stumbled across the Shapes3D library and the BezTube class, which seems to be perfect for drawing the curve with actual geometry. However, what I can't seem to find in any examples or documentation is whether I can easily animate the BezTube along the bezier curve (in much the same way that I use bezierPoint to get positions along the curve using a number between 0 and 1) without adjusting the control points along its entire length.

If anyone has experience in this, please let me know.

Activity Trend