Hi there,

it would be great if someone could help me understand some facts about cl-eye in combination with CCV.

my goal is to use ccv with many (>2) ps3 cams. I understand that you’ll need the non-free packs to use more than 2 cams. However, in the ccv forums i read that cl-eye will not support more than 2 cams in combination with ccv and the stitching functionality. is it so?

also, after checking the source of CCV, it seems no stitching is handled there, so i assume all of it is handled in the driver itself? Does the driver / camera-options-window check if it’s opened from ccv and displays the stitching options only in that case?

More stitching questions… it seems in CCV/PS3 you can stitch cameras horizontally only. In terms of optics and maximizing tracking resolution it would make more sense to stitch the cams vertically (resolution 640*960) in my case. is there a way to achieve such a setup with the current stitching implementation?

thank you very much!

Hello,

The PS3Eye input component you are talking about was developed especially for Windows version of CCV. It includes limited camera settings and horizontal camera stitching and supports up to two cameras. The settings dialog is built-in and it is opened using a single function call from within CCV. Because this is a generic input component, it could be used in other programs as well.
This was a separate project that I developed a while ago and it is not connected or dependent on a current CL-Eye codebase.

However, using the latest CL-Eye Platform SDK you can fully control all of the camera parameters and accomplish unlimited camera stitching functionality in any configuration (horizontal, vertical, grid, etc) and expose it to the CCV or any other software as a single image source. A custom development is needed to create such a component of course.

AlexP

Hi AlexP,

Quick question from a noob programmer.

I f I buy the quad pack I’m not directly ably to use all the 4 cams in ccv and stich them together?

As mentioned earlier, CCV only supports two cameras as of now. To support more cameras, you would need to write a new input component using CLEye SDK to take advantage of its multi-camera support.

Has anyone done this? Stitched together multiple cameras and sent it into CCV by writing their own custom component? I have scoured all the usual forums and blogs and haven’t found anyone that is actually using this method.

It seems to me that it is the best way to go. In theory, at least.

Also, if someone has written this custom piece of software, would anyone else be able to use it? I ask because I don’t have a lot of C++ knowledge, but need to stitch together several cameras for a project. I will struggle and hack my way through the process if I have to, but if this is a good solution like it seems, why hasn’t someone else done it and released the code?

I don’t mean to sound lazy. I just know that my area of expertise doesn’t lie in this area of code writing and that someone smarter than me would have an easier time of it for sure. I would also be great to put in the gallery of videos that you have along with the OctalPS3 video.

Anyway, if anyone has knowledge of an implemented solution, please post it here.