* first commit
* turned C from input to constant; force C constant in impl; better handling 0d/1d cases
* integrate with gemm from ficus nn
* fix const inputs
* adjust threshold for int8 tryQuantize
* adjust threshold for int8 quantized 2
* support batched gemm and matmul; tune threshold for rcnn_ilsvrc13; update googlenet
* add gemm perf against innerproduct
* add perf tests for innerproduct with bias
* fix perf
* add memset
* renamings for next step
* add dedicated perf gemm
* add innerproduct in perf_gemm
* remove gemm and innerproduct perf tests from perf_layer
* add perf cases for vit sizes; prepack constants
* remove batched gemm; fix wrong trans; optimize KC
* remove prepacking for const A; several fixes for const B prepacking
* add todos and gemm expression
* add optimized branch for avx/avx2
* trigger build
* update macros and signature
* update signature
* fix macro
* fix bugs for neon aarch64 & x64
* add backends: cuda, cann, inf_ngraph and vkcom
* fix cuda backend
* test commit for cuda
* test cuda backend
* remove debug message from cuda backend
* use cpu dispatcher
* fix neon macro undef in dispatcher
* fix dispatcher
* fix inner kernel for neon aarch64
* fix compiling issue on armv7; try fixing accuracy issue on other platforms
* broadcast C with beta multiplied; improve func namings
* fix bug for avx and avx2
* put all platform-specific kernels in dispatcher
* fix typos
* attempt to fix compile issues on x64
* run old gemm when neon, avx, avx2 are all not available; add kernel for armv7 neon
* fix typo
* quick fix: add macros for pack4
* quick fix: use vmlaq_f32 for armv7
* quick fix for missing macro of fast gemm pack f32 4
* disable conformance tests when optimized branches are not supported
* disable perf tests when optimized branches are not supported
* decouple cv_try_neon and cv_neon_aarch64
* drop googlenet_2023; add fastGemmBatched
* fix step in fastGemmBatched
* cpu: fix initialization ofb; gpu: support batch
* quick followup fix for cuda
* add default kernels
* quick followup fix to avoid macro redef
* optmized kernels for lasx
* resolve mis-alignment; remove comments
* tune performance for x64 platform
* tune performance for neon aarch64
* tune for armv7
* comment time consuming tests
* quick follow-up fix
VIT track(gsoc realtime object tracking model) #24201
Vit tracker(vision transformer tracker) is a much better model for real-time object tracking. Vit tracker can achieve speeds exceeding nanotrack by 20% in single-threaded mode with ARM chip, and the advantage becomes even more pronounced in multi-threaded mode. In addition, on the dataset, vit tracker demonstrates better performance compared to nanotrack. Moreover, vit trackerprovides confidence values during the tracking process, which can be used to determine if the tracking is currently lost.
opencv_zoo: https://github.com/opencv/opencv_zoo/pull/194
opencv_extra: [https://github.com/opencv/opencv_extra/pull/1088](https://github.com/opencv/opencv_extra/pull/1088)
# Performance comparison is as follows:
NOTE: The speed below is tested by **onnxruntime** because opencv has poor support for the transformer architecture for now.
ONNX speed test on ARM platform(apple M2)(ms):
| thread nums | 1| 2| 3| 4|
|--------|--------|--------|--------|--------|
| nanotrack| 5.25| 4.86| 4.72| 4.49|
| vit tracker| 4.18| 2.41| 1.97| **1.46 (3X)**|
ONNX speed test on x86 platform(intel i3 10105)(ms):
| thread nums | 1| 2| 3| 4|
|--------|--------|--------|--------|--------|
| nanotrack|3.20|2.75|2.46|2.55|
| vit tracker|3.84|2.37|2.10|2.01|
opencv speed test on x86 platform(intel i3 10105)(ms):
| thread nums | 1| 2| 3| 4|
|--------|--------|--------|--------|--------|
| vit tracker|31.3|31.4|31.4|31.4|
preformance test on lasot dataset(AUC is the most important data. Higher AUC means better tracker):
|LASOT | AUC| P| Pnorm|
|--------|--------|--------|--------|
| nanotrack| 46.8| 45.0| 43.3|
| vit tracker| 48.6| 44.8| 54.7|
[https://youtu.be/MJiPnu1ZQRI](https://youtu.be/MJiPnu1ZQRI)
In target tracking tasks, the score is an important indicator that can indicate whether the current target is lost. In the video, vit tracker can track the target and display the current score in the upper left corner of the video. When the target is lost, the score drops significantly. While nanotrack will only return 0.9 score in any situation, so that we cannot determine whether the target is lost.
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake
Rewrite Universal Intrinsic code: ImgProc (CV_SIMD_WIDTH related Part) #24166
Related PR: #24058, #24132. The goal of this series of PRs is to modify the SIMD code blocks in the opencv/modules/imgproc folder by using the new Universal Intrinsic API.
The modification of this PR mainly focuses on the code that uses the `CV_SIMD_WIDTH` macro. This macro is sometimes used for loop tail processing, such as `box_filter.simd.hpp` and `morph.simd.hpp`.
```cpp
#if CV_SIMD
int i = 0;
for (i < n - v_uint16::nlanes; i += v_uint16::nlanes) {
// some universal intrinsic code
// e.g. v_uint16...
}
#if CV_SIMD_WIDTH > 16
for (i < n - v_uint16x8::nlanes; i += v_uint16x8::nlanes) {
// handle loop tail by 128 bit SIMD
// e.g. v_uint16x8
}
#endif //CV_SIMD_WIDTH
#endif// CV_SIMD
```
The main contradiction is that the variable-length Universal Intrinsic backend cannot use 128bit fixed-length data structures. Therefore, this PR uses the scalar loop to handle the loop tail.
This PR is marked as draft because the modification of the `box_filter.simd.hpp` file caused a compilation error. The cause of the error is initially believed to be due to an internal error in the GCC compiler.
```bash
box_filter.simd.hpp:1162:5: internal compiler error: Segmentation fault
1162 | }
| ^
0xe03883 crash_signal
/wafer/share/gcc/gcc/toplev.cc:314
0x7ff261c4251f ???
./signal/../sysdeps/unix/sysv/linux/x86_64/libc_sigaction.c:0
0x6bde48 hash_set<rtl_ssa::set_info*, false, default_hash_traits<rtl_ssa::set_info*> >::iterator::operator*()
/wafer/share/gcc/gcc/hash-set.h:125
0x6bde48 extract_single_source
/wafer/share/gcc/gcc/config/riscv/riscv-vsetvl.cc:1184
0x6bde48 extract_single_source
/wafer/share/gcc/gcc/config/riscv/riscv-vsetvl.cc:1174
0x119ad9e pass_vsetvl::propagate_avl() const
/wafer/share/gcc/gcc/config/riscv/riscv-vsetvl.cc:4087
0x119ceaf pass_vsetvl::execute(function*)
/wafer/share/gcc/gcc/config/riscv/riscv-vsetvl.cc:4344
0x119ceaf pass_vsetvl::execute(function*)
/wafer/share/gcc/gcc/config/riscv/riscv-vsetvl.cc:4325
Please submit a full bug report, with preprocessed source (by using -freport-bug).
Please include the complete backtrace with any bug report.
```
This PR can be compiled with Clang 16, and `opencv_test_imgproc` is passed on QEMU.
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [ ] I agree to contribute to the project under Apache 2 License.
- [ ] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [ ] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake
Python: support tuple src for cv::add()/subtract()/... #24074
fix https://github.com/opencv/opencv/issues/24057
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [ x The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake
Rewrite Universal Intrinsic code by using new API: ImgProc module Part 2 #24132
The goal of this series of PRs is to modify the SIMD code blocks guarded by CV_SIMD macro in the opencv/modules/imgproc folder: rewrite them by using the new Universal Intrinsic API.
This is the second part of the modification to the Imgproc module ( Part 1: #24058 ), And I tested this patch on RVV (QEMU) and AVX devices, `opencv_test_imgproc` is passed.
The patch is partially auto-generated by using the [rewriter](https://github.com/hanliutong/rewriter).
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [ ] I agree to contribute to the project under Apache 2 License.
- [ ] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [ ] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake
In the previous code, there was a memory leak issue where the
previously allocated memory was not freed upon a failed realloc
operation. This commit addresses the problem by releasing the old
memory before setting the pointer to NULL in case of a realloc failure.
This ensures that memory is properly managed and avoids potential
memory leaks.
Skip test on SkipTestException at fixture's constructor (version 2) #24250
### Pull Request Readiness Checklist
Another version of https://github.com/opencv/opencv/pull/24186 (reverted by https://github.com/opencv/opencv/pull/24223). Current implementation cannot handle skip exception at `static void SetUpTestCase` but works on `virtual void SetUp`.
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake
Merge pull request #24274 from vrabaud:webp_1.3.2
Bump libwebp to 1.3.2 #24274
This is version [c1ffd9a](c1ffd9ac75)
It is 1.3.2 with a few patches that were made right after to help compilation.
No need for patches on the OpenCV side!
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
Rewrite Universal Intrinsic code by using new API: ImgProc module. #24058
The goal of this series of PRs is to modify the SIMD code blocks guarded by CV_SIMD macro in the `opencv/modules/imgproc` folder: rewrite them by using the new Universal Intrinsic API.
For easier review, this PR includes a part of the rewritten code, and another part will be brought in the next PR (coming soon). I tested this patch on RVV (QEMU) and AVX devices, `opencv_test_imgproc` is passed.
The patch is partially auto-generated by using the [rewriter](https://github.com/hanliutong/rewriter), related PR https://github.com/opencv/opencv/pull/23885 and https://github.com/opencv/opencv/pull/23980.
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [ ] I agree to contribute to the project under Apache 2 License.
- [ ] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [ ] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake
Fix undefined behavior arithmetic in copyMakeBorder and adjustROI. #24260
This is due to the undefined: negative int multiplied by size_t pointer increment.
To test, compile with:
```
mkdir build
cd build
cmake ../ -DCMAKE_C_FLAGS_INIT="-fsanitize=undefined" -DCMAKE_CXX_FLAGS_INIT="-fsanitize=undefined" -DCMAKE_C_COMPILER="/usr/bin/clang" -DCMAKE_CXX_COMPILER="/usr/bin/clang++" -DCMAKE_SHARED_LINKER_FLAGS="-fsanitize=undefined -lubsan"
```
And run:
```
make -j opencv_test_core && ./bin/opencv_test_core --gtest_filter=*UndefinedBehavior*
```
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [x] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake
Added default dimension value to tensorflow ArgMax and ArgMin layers #24266
Added default dimension value to tensorflow ArgMax and ArgMin layers.
Added exception when accessing layer's input with out of range index.
Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=48452
Modify the outputVideoFormat after changing the output format in MSMF backend #24142
After changing the output format, need to modify the outputVideoFormat, otherwise the outputVideoFormat is always CV_CAP_MODE_BGR, and an error will occur when converting the format in retrieveVideoFrame(), and will always enter "case CV_CAP_MODE_BGR:" process.
### Pull Request Readiness Checklist
See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request
- [ ] I agree to contribute to the project under Apache 2 License.
- [ ] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [ ] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake
Co-authored-by: 李龙 <lilong@sobey.com>
Dmitry Kurtaev
Alex
Alexander Smorkalov
Yuantao Feng
lpylpy0514
HAN Liutong
Kumataro
Kuan-Wei Chiu
Stefan Dragnev
Vincent Rabaud
Yuriy Chernyshov
Alexander Lyulkov
alexlyulkov
FlyinTeller
Alexander Alekhin
Alexander Smorkalov
jason_w
beanjoy