Quoted from "mstoss"

Quoted from "City3d"

Thank`s for the info. The problem, untill that, is that i`m shooting all images with a fish eye lens and same for the video. How can i crop or mask or how can i hide the black areas from the fish eye video in order to overlap on the pano?

Well, I do use the same lens for pano and video as well now.

My approach is, inspired by part of the discussion in this thread here:

Example/Tutorial - Extract a Distorted Hotspot from any kind of Pano Image (with perfect pixel alignment) You should read some posts there to get an idea.

Basically, I use ptgui and templates and batches to produce new frames out of the existing video:

1.Stitch your panoramic as usual
2. Export frames out of your video
3. Load one frame into your existing panorama and align. Resolution will differ, but that is no problem for ptgui.
4. Use this project as a template for batch exporting all the frames according to Klaus' perfect pixel alignment instructions Example/Tutorial - Extract a Distorted Hotspot from any kind of Pano Image (with perfect pixel alignment) By doing so, you will zoom in into the video zone only, don't worry about black areas outside the lens circle then.
5. Reassemble the frames to a video file.

(You can add a step to color match panorama and video if need be, there seem to be different ways to do that, I use a Photoshop action to batch the frames.)

Hope this is of some help

Michael

Quoted from "mstoss"

Quoted from "City3d"

With the video, i have broke it in frames (as you told) and for eg i`ve got 300 frames. Untill here was easy. But i don`t understand when you are saying that i`ll have to load one frame into existing panorama and align. From what i have read, some are saying that i have to import in ptgui only a frame from the video and crop it, than use batch builder with the template for all the frames. I don`t understand exactly this stepts.

Well, thats pretty much the same, I guess

You will need to:

1. align the video "frame image" in the same way you would with one of the regular fisheye images.
So load the 360° panoramic image in PTGUI first, then add the frame image. Since your camera is capturing images and videos with different dimensions, you will find, that the frame image is much smaller due to the limited video resolution. So in the lens settings tab check the "use individual lens parameters" for the frame image and roughly align it in the editor window. Then set your control points and optimize (only the frame image!). The frame image should fit the panoramic then.

2. export the frame-to-be out of the frame image.
In the panorama editor you can then zoom in to the desired video area, and export as tiff or jpg. Write down the parameters in the image parameters tab according to the perfect pixel alignment thread linked above. You could now replace the frame image no. 1 by no.2 and export again and repeat 300 times... You'll understand that it is much better to use the batch procedure offered by PTGUI for this purpose. So save the project and use as batch template for the rest of the frames (have a look at the ptgui help, if in doubt).

3. reassemble the video.
When all the frames are processed, you can build a movie file out of them again.

4. embed the video file
This is done analog to the perfect pixel alignment thread again, i.e. with the calc_pos_from_hfov_yaw_pitch_roll action call and the values which you wrote down in step 2

Hope this helps

Michael