We have been able to use the NVIDIAOpticalFlowSDK and generate good flow data but we are running into issues when attempting to run optical flow on frames that have already been allocated on the GPU. Are there any examples or details for using optical flow w/ existing on-device frames versus uploading new frames? We tried various ways such as mimicking the NV_OF_EXECUTE_INPUT_PARAMS, making modified OF buffers that use the pointers for the existing frames, etc. but we are not seeing valid output that’s even remotely close to what works when re-uploading and running through the SDK code.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Turing optical flow emulator that can run on CPU | 2 | 357 | May 2, 2023 | |
Best Practices for Implementing Optical Flow in Real-time Video Processing | 0 | 615 | April 15, 2023 | |
OpenCV and the latest OpticalFlow | 0 | 496 | November 25, 2019 | |
About the algorithm of Turing Optical Flow | 0 | 818 | December 1, 2019 | |
Optical Flow on NVIDIA PARKER | 1 | 540 | June 18, 2019 | |
Are there any examples about video optical flow SDK | 0 | 418 | August 10, 2019 | |
Accelerated Motion Processing Brought to Vulkan with the NVIDIA Optical Flow SDK | 0 | 387 | April 13, 2023 | |
Optical Flow SDK 5.0 released | 3 | 1797 | October 23, 2024 | |
Plan to upgrade NVIDIA OpticalFlow SDK in GitHub | 2 | 779 | November 21, 2019 | |
Return error when i use NV Optical Flow API | 4 | 530 | October 12, 2021 |