Queries on pix4d mapper

Hi i have following queries regarding pix4d version 4.7.5

  1. In pix4d version 4.7.5, what is the maximum number of RGB and multi spectral images that can be processed as a single project…size of each RGB image is 33 MB, and size each band in multi spectral camera us 6 MB, iam using sux band multi spectral camera…my workstation configuration is processor @ 2.6 G hz, 28 core, 128 GB RAM…with this configuration what is the maximum number of images i can process as a single project for RGB and multi spectral.
  2. is it possible to generate individual orthophotos for each frame in project or we can generate only orthomosaic?
  3. can i extract stereo map for RGB dats,.i.e 3D vector extraction of features viz building and other features from pix4d project ( ver 4.7.5)
    Can any one suggest me regarding above points.
    Thank you & Regards
    Narender

Hi @narender.bommineni,

Thank you for your request:

  1. There isn’t a specific limit on the number of images as this might depend on several factors.
    For very large projects (1000 - 2000 images at 14 MP) we recommend this computer specifications 64 GB RAM, 120 GB SSD Free Space. If your specifications are higher, you can use more images.
  2. Could you please provide more information about this request? Are you requesting to have every image transformed in an orthophoto? If this is the case, PIX4Dmapper is not creating this kind of output (also, it will not be possible to create these images for images at the border of the project). If you need an orthoimage of a region smaller than the entire project, you can use the Processing area feature in PIX4Dmapper as described here:
    https://support.pix4d.com/hc/en-us/articles/202560179-How-to-use-a-processing-area-in-Pix4Dmapper
  3. Could you please provide more information about this request? Here is a list of outputs that PIX4Dmapper creates: https://support.pix4d.com/hc/en-us/articles/202558499-Pix4D-outputs-with-other-software-by-output
    To vectorize in PIX4Dmapper you can draw polylines and surfaces:
    https://support.pix4d.com/hc/en-us/articles/202560309-How-to-draw-a-Polyline-in-the-rayCloud
    https://support.pix4d.com/hc/en-us/articles/202560269-How-to-draw-a-Surface-in-the-rayCloud
    For more detailed vectorization options, you can give PIX4Dsurvey a try: https://www.pix4d.com/product/pix4dsurvey
    https://support.pix4d.com/hc/en-us/articles/360033317432-Vectorization-PIX4Dsurvey

Thank you sir for your response.

  1. rgb camera is having 42 mp resolution and each image size is around 33 MB… is it possible to create single project which has images of 70000 to 100000.
    My computer work station is 128 GB RAM, 28 cores, 2.9 GHZ
  2. is it possible to generate ortho photo for each frame in project…For example if project has 20000 rgb images is it possible to create 20000 orthophotos( that means orthophoto for each frame).I know we can generate orthomosaic, but iam interested in orthophoto generation
  3. iam interested in 3d vector capture ( stereo capture by seeing through 3d glasses). My interest is to generate capture top of building in stereo mode. Is it possible to do stereo mapping in pix4d mapper or is it we need to export to other software for 3d vectorisation.
    Thank you.
    Regards
    Narender

I also have a 42mp camera. I successfully performed a single project with about 11,000 photos. I tried another project with 22,000 photos that failed. You’ll need to use Pix4DMatic for these very large projects.

Also, your workstation is fine for medium size projects, but not for very large ones. It may work, but will be slow. I would recommend an AMD Threadripper Pro 5995wx and at least 256gb of memory.

Check out Puget Systems Pix4DMapper benchmarks. Intel Xeon processors perform poorly compared to AMD Threadripper Pro.

hi
–thank you for your response to my query.
i have following 2 queries regarding pix4d processing

  1. is it possible to generate orthophoto for each frame in project…For example if project has 20000 rgb images is it possible to create 20000 orthophotos( that means orthophoto for each frame).I know we can generate orthomosaic, but iam interested in orthophoto generation
  2. iam interested in 3d vector capture ( stereo capture by seeing through 3d glasses). My interest is to generate capture top of building in stereo mode. Is it possible to do stereo mapping in pix4d mapper or is it we need to export to other software for 3d vectorisation.
    –hoping reply for above queries
    –thank you for your time

Regards

Hi @narender.bommineni

  1. No, it is not possible to create orthophotos, just an orthomosaic. You can crop the orthomosaic with a third-party application if you need a smaller area.
    However, it is possible to save undistorted images, considering the optimized distortion parameters. You can find additional information in this article:

https://support.pix4d.com/hc/en-us/articles/202557929-Menu-Process-Save-Undistorted-Images

  1. These are the outputs created by PIX4Dmapper: https://support.pix4d.com/hc/en-us/articles/202558499-Pix4D-outputs-with-other-software-by-output. You can have a simple vectorization tool using pix4dmapper (Draw a polyline, Draw a surface). For more detailed vectorization options, you can give PIX4Dsurvey a try: https://www.pix4d.com/product/pix4dsurvey
    https://support.pix4d.com/hc/en-us/articles/360033317432-Vectorization-PIX4Dsurvey.
    Once you have the 3D model, you will need third-party software to convert it into a file readable by 3d glasses.

Let me know if this information helps you or if there is anything else I can help you with.