- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
We are developing application which using OpenCV on Java with org.bytedeco platform:
<dependency>
<groupId>org.bytedeco</groupId>
<artifactId>opencv-platform-gpu</artifactId>
<version>4.1.2-1.5.2</version>
</dependency>
And we want to use matching algorythms with GPU.
The problem is that when we creatint train collection of imwges descriptors with number on GpuMat we get memory allocation error for only 500+ GpuMats which summary size is about 40Mb:
gpu_mat.cu:116: error: (-217:Gpu API call) out of memory in function '<unnamed>::DefaultAllocator::allocate'GPU used for testing Nvidia 750 with 2GbHow we can create more GpuMats to load more images for matching?
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi gicsoft,
Are you using OpenVINO in any part of your program? It seems that this is an OpenCV or CUDA API issue.
I would recommend you to contact Nvidia or the following OpenCV resources:
Regards,
Rizal
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank you for your question. If you need any additional information from Intel, please submit a new question as Intel is no longer monitoring this thread.
Regards,
Rizal

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page