-
Notifications
You must be signed in to change notification settings - Fork 4.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Disable cameras after fetching images, projection matrix #2881
Conversation
I just tried this and this improved my camera API retrieval FPS a lot! With 6 camera requests (2 cameras with each RGB/depth/segmentation) at 640x480 I can get 30/40FPS. Were previously it would often cripple to below 10FPS. I'll do some more testing but so far really nice result. |
Wow, pretty awesome, didn't expect that it would have that much of an effect, the main improvement is apparent when there's a high res camera used infrequently. Thought that this would actually slow down the normal continuous usage a tiny bit. Related to this, I do hope to get some work done on the other faster_img_capture PR over the next week, after some deadlines this weekend. I think as a last resort might even need to go for separate calls for Vulkan and OpenGL, DirectX. Any suggestions would be great |
…rosoft#2881)" Running the Image API seems to trigger exceptions This reverts commit ba4be15.
…rosoft#2881)" Running the Image API continuously seems to trigger exceptions This reverts commit ba4be15.
Follow up, this seems to be causing crashes when using the API continuously, strange that it didn't appear with my testing and others as well. I've opened #3039 which reverts this, but again needs testing to confirm |
…rosoft#2881)" Running the Image API continuously seems to trigger exceptions This reverts commit ba4be15
Replaces #2465 by doing that internally
Before -
After -
Not sure as to what is causing the still present increase in time, but still much better than earlier
Tested whether disabling camera affects subwindow, no effect
Haven't tested the projection matrix one though
Closes #1766, closes #2389, closes #2464