PIL Image.fromstring from PyOpengl buffer has the wrong size

Question

I use PyOpenGL to draw a 2D Image. Then I want to use the Python Imaging Library (PIL) to store this image to disk. I use GLUT to display the image which works perfectly. But when I use PIL to store the image it extracts the wrong clipping. It has the wrong size.

Here is a minimal example which reproduces the effect and I also attach the output to make it more clear without running some code.

from OpenGL.GL import *
from OpenGL.GLUT import *
from PIL import Image

width, height = 640, 480

def DrawStuff():
    poly1 = [(0,0), (640,0), (0,480)]
    color = (0.5, 0.4, 0.3, 0.8)
    glClear(GL_COLOR_BUFFER_BIT)
    glPushMatrix()
    glLineWidth(5.0)

    glColor4f(*color)
    glBegin(GL_POLYGON)
    glVertex2f(poly1[0][0], poly1[0][1])
    glVertex2f(poly1[1][0], poly1[1][1])
    glVertex2f(poly1[2][0], poly1[2][1])    
    glVertex2f(poly1[0][0], poly1[0][1])
    glEnd() # GL_POLYGON

    glPopMatrix()
    glPixelStorei(GL_PACK_ALIGNMENT, 1)
    data = glReadPixels(0, 0, width, height, GL_RGBA, GL_UNSIGNED_BYTE)
    image = Image.fromstring("RGBA", (width, height), data)
    image.show()
    image.save('out.png', 'PNG')
    glutSwapBuffers()


# glut initialization
glutInit(sys.argv)
glutInitDisplayMode(GLUT_DOUBLE | GLUT_RGBA)
glutCreateWindow("Draw Polygons")
glutInitWindowSize(width, height)

# set the function to draw
glutDisplayFunc(DrawStuff)

# enable the alpha blending
glEnable(GL_BLEND)
glBlendFunc (GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA);

# prepare for 2D drawing
glMatrixMode(GL_PROJECTION)
glLoadIdentity()
glOrtho(0, width, height, 0, 0, 1)
glDisable(GL_DEPTH_TEST)
glMatrixMode(GL_MODELVIEW)

# start the mainloop
glutMainLoop ()

this is how it looks int the GLUT window and how it is supposed to look like

and this is how the saved image looks like

score 1 · Accepted Answer · edited May 23 '17 at 10:34

I managed to solve my own Problem.

First I tried the following solution which might also help people with related problems: solution1

But then, through extensive trial and error, I found that the solution is much simpler.

I simply had to swap two lines from:

glutCreateWindow("Draw Polygons")
glutInitWindowSize(width, height)

to

glutInitWindowSize(width, height)
glutCreateWindow("Draw Polygons")

Apparently the size has to be set before the window

score 0 · Answer 2 · answered Aug 18 '11 at 13:19

0

You should consider that in OpenGL the coordinate system starts at different place than in PIL. Look at this.

answered Aug 18 '11 at 13:19

Mariy

5,746
4
40
57

thanks for sour help. But his was not the problem. I found the solution but cannot post it as an answer because I don't have enough reputation. I'll edit the question. – dedan Aug 18 '11 at 13:56
1

This is a shining example of why answers shouldn't rely on links to explain. Note the dead link. It's a shame as the question (and potentially this answer) are still relevant. – ephsmith Aug 22 '12 at 18:18

PIL Image.fromstring from PyOpengl buffer has the wrong size

2 Answers2