Gpu frame comparison 96 - #111
Merged
Merged
Conversation
- Move frame comparison operations from CPU to GPU - Created GPUFrameChangeDetector with OpenCV CUDA functions - Implemented GPU-based MSE, SSIM, and optical flow calculations - Added gpu_frames_skipping_pipeline.py with GPU frame processing - Uses cv2.cuda.resize, cv2.cuda.absdiff, cv2.cuda.multiply for GPU ops - Downloads to CPU only for final calculations (mean, std) - Eliminates GPU→CPU→GPU copies for frame processing - Maintains same thresholds and logic as CPU version
Co-authored-by: Sarah Gershuni <sarah556726@gmail.com> Co-authored-by: Chani Orlinski <c9992946@gmail.com>
Co-authored-by: Sarah Gershuni <sarah556726@gmail.com> Co-authored-by: Chani Orlinski <c9992946@gmail.com>
- Implemented CUDA-EGL GPU buffer access from DeepStream - Added C++ pybind11 wrapper for GPU frame processing - Integrated MSE, SSIM, and Optical Flow detection algorithms - Created GPU-based frame skipping pipeline for DeepStream - Added Makefile for building C++ extensions with CUDA support
- Add C++ GPU frame change detector with CUDA-EGL integration - Add pybind11 wrapper for DeepStream GPU buffer access - Update pipeline to use GPU-only frame processing - Fix Makefile to compile correct GPU modules - Implement same logic as CPU version but entirely on GPU
- Updated GPU thresholds: MSE<10.0, SSIM>0.995, FLOW<0.05 - Achieved 7/50 frame processing (86% skip rate) - Removed deprecated gpu_buffer files - Updated gpu_probe with improved frame processing
lyuzinmaxim
reviewed
Nov 10, 2025
lyuzinmaxim
reviewed
Nov 10, 2025
|
|
||
| stats = {"total": 0, "skipped": 0, "processed": 0} | ||
|
|
||
| def get_jetson_frame(gst_buffer, batch_id): |
Collaborator
There was a problem hiding this comment.
are you using this function? I am a bit confused since to me it seems like it's not called anywhere
lyuzinmaxim
reviewed
Nov 10, 2025
| print(f" Skip Ratio: {skip_ratio:.1%}") | ||
|
|
||
| if __name__ == "__main__": | ||
| run_jetson_pipeline() No newline at end of file |
Co-authored-by: Sarah Gershuni <sarah556726@gmail.com> Co-authored-by: Chani Orlinski <c9992946@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
GPU-Accelerated Frame Change Detection for DeepStream
Implements intelligent frame skipping using GPU-only processing to optimize DeepStream pipeline performance.
Key Features
Closes #96