Open JV-X opened 2 months ago
Not sure what you mean. Do you need an ortho projection of the entire scene?
Not sure what you mean. Do you need an ortho projection of the entire scene?
Yes, it seems that ortho projection is exactly what I need. So, how can I obtain the ortho projection of the entire scene from a top-down view from the MVS file? thanks for your time.
I implemented some ortho projection functionality, but currently there is not a full pipeline that generates a complete orthoimage. I'll add it when I find some time.
I implemented some ortho projection functionality, but currently there is not a full pipeline that generates a complete orthoimage. I'll add it when I find some time.
When will this feature likely be available? Should I look for alternative methods to obtain the orthographic projection?
and one more question, are the images you project colored? I've found some examples online where projections are generated from PLY files, but the projected images are not colored.
thanks a lot.
I'll add it when I find some free time. Pls try open drone map project. It has an implementation of Ortho map projection
I'll add it when I find some free time. Pls try open drone map project. It has an implementation of Ortho map projection
thanks so much, I'll have a try.
I have tried OpenDroneMap (https://github.com/OpenDroneMap/ODM), but the images it generates are like the ones below, for me. it's too disordered to be used for human pose recognition. Do you have any other methods to try?
thanks so much.
ODM is designed for aerial scenes no other suggestion for now
ODM is designed for aerial scenes no other suggestion for now
okay, thanks for your reply.
We would like to obtain a projection image with the camera positioned directly above, based on the generated dense point cloud MVS file, for use in MediaPipe's human pose recognition(mediapipe: https://mediapipe-studio.webapps.google.com/demo/pose_landmarker). Is this idea feasible? Is there a method to obtain a projection image from an dense point cloud MVS file?
thanks for your time.