Smooth scene zoom out

[DVangelis]

Hello,
The last couple days I experiment with plotting isosurfaces of tiff 3D stacks.
I save the plt in every interation so later I can make a gif of the all plotters.
I have two questions first it seems that the plt for the first iterations zoom in and plots only in regions where it can find isosurfaces. And as the isosurface of interest grows with iterations it zooms out.
How I can select a plotter view with the dimensions of 3D stack and not of the isosurface in interest.
Secondly, I want to find a way to zoom out every couple iterations , I have found one solution but I dont know if it is the most eligible by changing the camera every few iterations. The details of cam I found it by turning offscreen= False and use the C to copy the cam dict.
Thanks in advance I post also a part of my solution to the above .

im= r" folder path"
f_nm=0
for idx,file in tqdm(enumerate(os.listdir(im))):
    if file.endswith('.tif'):
        image=glob(im+f'\{file}')
        img=tifffile.imread(image)
        if idx <=13:
            cam = dict(
            pos=(235.4933, 802.3488, 577.0033),
            focalPoint=(211.5927, 222.6740, 54.83333),
            viewup=(0.01202570, 0.6689701, -0.7431920),
            distance=780.5483,
            clippingRange=(246.7147, 1454.417),
        )
        elif idx>13 and idx<15:
            cam = dict(
            pos=(240.5125, 924.0806, 686.6590),
            focalPoint=(211.5927, 222.6740, 54.83333),
            viewup=(0.01202570, 0.6689701, -0.7431920),
            distance=944.4634,
            clippingRange=(408.9907, 1620.791),)
        elif idx>=15 and idx<20:
            cam = dict(
            pos=(246.5856, 1071.376, 819.3424),
            focalPoint=(211.5927, 222.6740, 54.83333),
            viewup=(0.01202570, 0.6689701, -0.7431920),
            distance=1142.801,
            clippingRange=(605.3446, 1822.104),
        )
        elif idx>=20 and idx<30:
            cam = dict(
            pos=(253.9341, 1249.603, 979.8893),
            focalPoint=(211.5927, 222.6740, 54.83333),
            viewup=(0.01202570, 0.6689701, -0.7431920),
            distance=1382.789,
            clippingRange=(842.9329, 2065.692),)

        elif idx>=30:
            cam = dict(
            pos=(262.8258, 1465.258, 1174.151),
            focalPoint=(211.5927, 222.6740, 54.83333),
            viewup=(0.01202570, 0.6689701, -0.7431920),
            distance=1673.175,
            clippingRange=(1130.415, 2360.433),
        )
      
        img = tifffile.imread(path)
        print(img.shape)
        vol = Volume(img).isosurface()
        vol.computeNormals().c('white')
        print('check0')
        plt = Plotter(N=1,bg='black',offscreen=True )
        plt.at(0)
        radius = vol.diagonalSize()/5  # need to specify it!
        plt.addAmbientOcclusion(radius)

        plt += vol.clone()
        plt.render()
        plt.addGlobalAxes(axtype=6)
        plt += Text2D('Experiment --'+"")

        axes = Axes(
            xrange=vol.xbounds(),
            yrange=vol.ybounds(),
            zrange=vol.zbounds(),
            xtitle="Z-axis",
            ytitle="Y-axis",
            ztitle="X-axis",
            xyGrid=False,
            axesLineWidth =2,
            zxGrid2= False,
            textScale= 2.0)
        plt += axes


        output=output_path+f"/stack_{f_nm}.png"
        plt.show(camera=cam)


        plt.screenshot(output)
        f_nm += 1     `

[marcomusy]

Uhm, while that's ok, I would experiment with using one single Plotter outside of the loop, and update (add()/remove()) objects while inside the loop.
You will need to call show() outside the loop once to setup the viewer.
You can also use moveCamera() to move it smoothly.

[DVangelis]

Ok with objects you mean the mesh that i create from each tif file correct ?
So to use to plotters
one will be a dump plotter , plt1= Plotter()
and I will cal plt1.show() outside the loop
and a second one as it is
with moveCamera() do I use camstart my first cam that I have in my code and as camstop the last one
and fraction it takes a number between 0-1 of how fast I want this transition to be ?
Last thing I know the voxel size of my tif stacks and I want to use it as scale in xrange,yrange,zrange how I can do that ?

[DVangelis]

I managed to keep the same scene from all frames.
But I am not able to use the moveCamera() I added to Plotter outside of the loop as you can see but does not seem to have any effect to the plt . Am I missing something ?

`f_nm=10
plt = Plotter(N=1,bg='black',resetcam=False,offscreen=True ).moveCamera(camstart=cam_start,camstop=cam_stop,fraction=1)
plt.addGlobalAxes(axtype=9)

axes = Axes(
xrange=(0,513),
yrange=(0,513),
zrange=(0,513),
xtitle="Y",
ytitle="Z",
ztitle="X",
xyGrid=False,
axesLineWidth =2,
zxGrid2= False,
textScale= 2.0)   
plt += axes
       
radius=146.18738348024212
plt.addAmbientOcclusion(radius)
plt.show()

for idx,file in tqdm(enumerate(os.listdir(im))):
    if idx>10:
        if file.endswith('.tif'):
            image=glob(im+f'\{file}')
            img=tifffile.imread(image)

            vol = Volume(img).isosurface()
            vol.computeNormals().c('white')
            radius = vol.diagonalSize()/5  # need to specify it!

            text =Text2D(''+f" time {tl['H-M-S'][idx]} (H:M:S)")
            plt.add(text)
            plt.add(vol.clone())

            output=r'f"/stack2_{f_nm}.png"
            plt.show()
            plt.screenshot(output)
            plt.remove(text)
            plt.remove(vol.clone())
            f_nm += 1  `

[marcomusy]

Good!
No, you need to put
plt.moveCamera(camstart=cam_start,camstop=cam_stop,fraction=idx/number_of_iterations)
inside the loop.

plt = Plotter(bg='black', resetcam=False, offscreen=True)

axes = Axes(
    xrange=(0,513),
    yrange=(0,513),
    zrange=(0,513),
    xtitle="Y",
    ytitle="Z",
    ztitle="X",
    xyGrid=False,
    axesLineWidth=2,
    textScale=2.0
)
plt += axes

text2d = Text2D(font="Calco")
plt += text2d

radius = 146.18738348024212
plt.addAmbientOcclusion(radius)
plt.show()

f_nm = 0
for idx,file in tqdm(enumerate(os.listdir(im))):
    if idx>10:
        if file.endswith('.tif'):
            # image = glob(im+f'\{file}') # you dont need this in principle
            # img = tifffile.imread(image)

            iso = Volume(file).isosurface()
            iso.computeNormals().c('white')

            text2d.text(f" time {tl['H-M-S'][idx]} (H:M:S)")

            plt.moveCamera(cam_start, cam_stop, idx/number_of_iterations)
            plt.add(iso) # this also renders the scene

            output = f"/stack2_{f_nm}.png"
            plt.screenshot(output)
            f_nm += 1

            plt.remove(iso)

[DVangelis]

Ok thanks
when I pass the file into iso = Volume(file).isosurface() it returns an index error index out of range
`
I managed to do use the idx/number_of_iterations
Thank you very much

[marcomusy]

when I pass the file into iso = Volume(file).isosurface() it returns an index error index out of range

can you post the exact error message?

[DVangelis]

Here it is

`---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_22220\4039960425.py in <module>
     31 #             img = tifffile.imread(image)
     32 
---> 33             iso = Volume(file).isosurface()
     34             iso.computeNormals().c('white')
     35 

~\anaconda3\lib\site-packages\vedo\volume.py in __init__(self, inputobj, c, alpha, alphaGradient, alphaUnit, mode, spacing, dims, origin, mapper)
    824         elif utils.isSequence(inputobj):
    825 
--> 826             if isinstance(inputobj[0], str) and ".bmp" in inputobj[0].lower():
    827                 # scan sequence of BMP files
    828                 ima = vtk.vtkImageAppend()

IndexError: list index out of range`

[marcomusy]

I don't understand .. I think you may be passing an empty list as file ? Maybe
iso = Volume(im+f'\{file}').isosurface()
would work?

[DVangelis]

Yes it worked and it seems that is faster.
Last thing.
how I can change the scale of the plt my voxel size is 2.75um so 1px of my volume is 2,75um
Should I use vedo.base scale and put s=2.75

[marcomusy]

You should be able to set the dataset spacing (size of the voxel) along the 3 axes separately with:
iso = Volume(im+f'\{file}', spacing=[2.75,2.75,2.75]).isosurface()