Alexander Alekhin

> opencv-python==4.6.0.66 There is no way to amend already released code (>1 year ago). Try to upgrade and verify the latest release. --- > `note: 'std::uintptr_t' is defined in header...

> [Abdurrahheem](https://github.com/Abdurrahheem) added this to the [5.x](https://github.com/opencv/opencv/milestone/57) milestone Why do you create a new milestone?

There are several options: - try to place `qt6_disable_unicode_defines` / `qt_disable_unicode_defines` in CMake for `opencv_world` target. Docs: https://doc.qt.io/qt-6/qt-disable-unicode-defines.html - try to `#undef UNICODE` in the beginning of trace.cpp file

There is still a problem in source code, so lets keep this issue open.

I believe yes. Try `testDarknetLayer(..., /*testBatchProcessing =*/ true)` in the mentioned test. Probably it should be targeted to 4.9.0+ milestone.

Please avoid posting of text information via screenshoots. It is ridiculous and blocks comparison of its output. As first step of investigation we should align build flags first (`CPU_BASELINE` at...

@ZJDATY Consider using OpenCV performance tests. There is test case for `yolov4-tiny` here: https://github.com/opencv/opencv/blob/4.8.0/modules/dnn/perf/perf_net.cpp#L242 Unfortunately used model in 4.8.0 and 4.5.5 are not the same (details #23008). Need to restore...

> @MaximMilashchenko [resolved branch conflicts](https://github.com/opencv/opencv/pull/25307/commits/4802254b933e298a06e8cd6e52ef57ca10bb74c0) `git rebase` should be done instead of merge. Internal merges should be avoided.

In C++11 it must be safe: > If multiple threads attempt to initialize the same static local variable concurrently, the initialization occurs exactly once (similar behavior can be obtained for...

relates #19670