Camera tracking is about to become a lot better with the new Ceres-based planar tracker.
Advertisement
You're blocking ads, which pay for BlenderNation. Read about other ways to support us.
About the Author
Bart Veldhuizen
I have a LONG history with Blender - I wrote some of the earliest Blender tutorials, worked for Not a Number and helped run the crowdfunding campaign that open sourced Blender (the first one on the internet!). I founded BlenderNation in 2006 and have been editing it every single day since then ;-) I also run the Blender Artists forum and I'm Head of Community at Sketchfab.
21 Comments
Nice!
Cool
these a build here http://www.graphicall.org/930
Thanks for taking the time to make such a nice presentation.
Good voice! straight to the point!
I had fun making this! I am debating making a longer one that describes how the solver works. It won't be interesting to everyone, but some might find it fascinating.
If you have time for it please do. That would certainly be interesting.
And good work! This will definitely make tracking easier.
Please do it. I saw the link to the google-code project and I was amazed by the things this solver does, and I'd really enjoy to know how it works...
I would be very interested to learn how the solver works. The affine fitting routines could be very useful for me to analyse atomic force microscopy images!
ah, tripod motion :-)
awesome! i cant wait to use it.,
Woah! It's really cool! Blender is becoming a real "swiss knife"! :)
OMG. Just working on a tracking project, wishing for more tracking features. Take a break, come to blendernation, and low and behold. This is so awesome. I just can't believe this team. I wish I could contribute code.
I know exactly what you mean. I'm making a game using Blender in my workflow, and there are times when I come here, I find just what I needed.
awesome ;-)
and you are helping us all by collection the awesome stuff for us :)
thank you!
Охрененно!
...it russian: "Wow+Cool+Awesome" — all together!Ё :D
Nice!
Looks great.
hmm... maybe this is a stupid question: but wouldn't the tracking in general be a lot less error prone if the real-world distance between markers were known? Not between 2 markers to get the dimensions right - but between groups of 3 markers to get their 3d relationship right. In most studio-environments measuring the distance between 3 points that are used as markers should be quite simple. And especially object-tracking that uses designated markers from the get-go could benefit a lot.
But maybe I'm just being silly.
This is not silly at all. If you have enough measurements, you can solve the camera position from a single frame.
Supporting survey data is planned.
I'm glad to hear that. But I have to admit that I was surprised that it isn't implemented yet. I thought that it would have been a lot easier than writing a solver that has to guesstimate all the distances. Less cool but easier. Therefore I thought that I was proposing an absolute no-go - the "goto"-of-camera-tracking. ;)
Thanks for the great work!