Corey Beaulieu Posted February 28, 2017 Share Posted February 28, 2017 Hi Everyone, I have a situation where I need to match a photo taken with an iPhone, an iPhone 7. I have done some requisite Googling and found a few options, but the thing that works best is a lens type of 3.99 with a Crop/Zoom factor of 7.21. Using this, I can locate the building with a reasonable approximation of where the photo was taken from. What is incorrect is the width of the building. Messing around with a few ideas, I landed on using a Lens Distortion amount and this is new territory for me. With nothing to go on I am using a Cubic Distortion of -0.721 which gives me a near correct width, but the height remains slightly off. Most things, however, are correct so I am really just looking to refine what I am working with. I want to ask if anyone has any advice here? What else can I try or what specifically do you know that I may use to help create a better match? The distortion keeps my camera on a Google map overlay that places my camera near correct and most things are working, but I'd like to know more. No advice is bad advice here so let 'er rip, but if I can make this match better I would welcome any thoughts or ideas. Thank you. Link to comment Share on other sites More sharing options...
Chris MacDonald Posted February 28, 2017 Share Posted February 28, 2017 Rather than try to work out crop factors, I usually google the sensor size of the device and set that as the film gate and then input whatever focal length (equivalent) the phone has in its specs. The computer then works out the crop factor/FOV etc. Link to comment Share on other sites More sharing options...
Scott Schroeder Posted February 28, 2017 Share Posted February 28, 2017 I just let it pass the look-see test. If it looks right, see then it's right. I've gone down the path of trying to make it a 100% match at the expense of doing actual work when my current 95% match will do just fine. Link to comment Share on other sites More sharing options...
pierreguay Posted April 28, 2017 Share Posted April 28, 2017 Corey, I was researching rhino to 3dsmax workflow and saw your post that led me to look at more of your posts so here I am.. A bit late but if it may comes up again. There are several ways you can accomplish this depending on how accurate your requirements are. (from most accurate to least..) Use a program like photomodeler, or any like these https://en.wikipedia.org/wiki/Comparison_of_photogrammetry_software You calibrate the lens on your camera no matter what it may be. You solve for the location of the camera with photomodeler and then export the camera to 3dsmax. Anything you put into the scene will be a perfect fit. Use sketchup's match photo option to build 3d model if box shaped elements are in your photo. Using this measure and estimate and build about 5 known 3d points in your photo you can use this info in 3dsmax camera match. Perspective match is another option. regards PIerre Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now