mirror of
https://github.com/opencv/opencv.git
synced 2026-07-29 07:13:02 +04:00
Vadim Pisarevsky
fd4491fcb1
The new graph engine has no FP16 execution path yet and runs FP32 on CPU regardless of the requested target. When an OpenCL FP16 target is requested, callers may pass CV_16F input types to Net::getFLOPS(), which then flow through tryInferShapes() -> Layer::getTypes() and get rejected (CV_16F is only allowed when preferableTarget == DNN_TARGET_OPENCL_FP16, which the new engine never sets). Normalize FP16 input types to FP32 in the mainGraph path of getFLOPS via a small filterFP16InputTypes() helper. This fixes the OCL_FP16 parameterization of the DNNTestNetwork perf tests (e.g. YuNet_320/640/1280) which crashed in getFLOPS. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
OpenCV: Open Source Computer Vision Library
Resources
- Homepage: https://opencv.org
- Courses: https://opencv.org/courses
- Docs: https://docs.opencv.org/5.x/
- Q&A forum: https://forum.opencv.org
- previous forum (read only): http://answers.opencv.org
- Issue tracking: https://github.com/opencv/opencv/issues
- Additional OpenCV functionality: https://github.com/opencv/opencv_contrib
- Donate to OpenCV: https://opencv.org/support/
Contributing
Please read the contribution guidelines before starting work on a pull request.
Summary of the guidelines:
- One pull request per issue;
- Choose the right base branch;
- Include tests and documentation;
- Clean up "oops" commits before submitting;
- Follow the coding style guide.
Additional Resources
- Submit your OpenCV-based project for inclusion in Community Friday on opencv.org
- Subscribe to the OpenCV YouTube Channel featuring OpenCV Live, an hour-long streaming show
- Follow OpenCV on LinkedIn for daily posts showing the state-of-the-art in computer vision & AI
- Apply to be an OpenCV Volunteer to help organize events and online campaigns as well as amplify them
- Follow OpenCV on Mastodon in the Fediverse
- Follow OpenCV on Twitter
- OpenCV.ai: Computer Vision and AI development services from the OpenCV team.