Forum

Author Topic: viewpoint in headless mode (-r) not set for PDF export  (Read 1833 times)

BenjaminG

  • Newbie
  • *
  • Posts: 16
    • View Profile
viewpoint in headless mode (-r) not set for PDF export
« on: April 03, 2018, 12:08:29 PM »
Hi,

When exporting 3D PDF with exportModel() in headless mode (-r) the model is completely zoomed out. But when using gui mode (--gui) everything is correct.

When I look at the PhotoScan.app.viewpoint values I get when using -r mode:
Code: [Select]
viewpoint.center : Vector([0.0, 0.0, 186.60254037844388])
viewpoint.coo    : Vector([0.0, 0.0, 0.0])
viewpoint.fov    : 30.0
viewpoint.height : 100
viewpoint.width  : 100
viewpoint.mag    : 1.0
viewpoint.rot    : Matrix([[1.0, 0.0, 0.0],
                           [0.0, 1.0, 0.0],
                           [0.0, 0.0, 1.0]])

And when using  --gui -r
Code: [Select]
viewpoint.center : Vector([-0.021930644007888106, -0.009145534891478346, 0.12193548540351035])
viewpoint.coo    : Vector([-0.021930644007888106, -0.009145534891478346, -0.1705619312323125])
viewpoint.fov    : 30.0
viewpoint.height : 577
viewpoint.width  : 1283
viewpoint.mag    : 3681.0467263858754
viewpoint.rot    : Matrix([[1.0, 0.0, 0.0],
                           [0.0, 1.0, 0.0],
                           [0.0, 0.0, 1.0]])

Then I tried modifying viewpoint:
Code: [Select]
new_viewpoint = PhotoScan.app.viewpoint
new_viewpoint.coo = PhotoScan.Vector([-0.021930644007888106, -0.009145534891478346, -0.1705619312323125])
new_viewpoint.fov = 30.0
new_viewpoint.mag = 3681.0467263858754
new_viewpoint.rot = PhotoScan.Matrix([[1.0, 0.0, 0.0],
                                      [0.0, 1.0, 0.0],
                                      [0.0, 0.0, 1.0]])
PhotoScan.app.viewpoint = new_viewpoint

But it has been completely ignored.

What I'm trying to get is the default viewpoint automatically calculated just like in --gui mode.

Thanks.

BenjaminG

  • Newbie
  • *
  • Posts: 16
    • View Profile
Re: viewpoint in headless mode (-r) not set for PDF export
« Reply #1 on: May 20, 2019, 05:24:35 PM »
Since I needed this feature I just made a workaround

Here it is if anyone have the same need:

Code: [Select]
def correct_pdf(in_file):

    chunk = Metashape.app.document.chunk
    chunk.resetRegion()

    T = chunk.transform.matrix

    # Center of region
    view_center = list(T.mulp(chunk.region.center))

    # get the region max size
    max_size = max(T.mulp(chunk.region.size))

    # Computre the distance to get a 30° view
    CO = (max_size / 2) / math.tan(math.radians(30/2))

    # Compute the center of the camera
    C2W = view_center
    view_center[2] += CO

    # Open the PDF and create a new one
    old_file = open(in_file, 'rb')
    new_file = open(in_file + '.newpdf', 'wb')
    for line in old_file:
        #  C2W
        if line.startswith(b'/C2W'):
            new_file.write(('/C2W [ -1 0 0 0 1 0 0 0 -1 ' +
                           str(round(C2W[0], 5)) + ' ' +
                           str(round(C2W[1], 5)) + ' ' +
                           str(round(C2W[2], 5)) + ' ]\n').encode('utf-8'))
        #  CO
        elif line.startswith(b'/CO'):
            new_file.write(('/CO ' + str(round(CO, 5)) + '\n').encode('utf-8'))
        else:
            new_file.write(line)
    old_file.close()
    new_file.close()