Try to enable Winograd by default in FP32 mode and disable it by default in FP16 mode #24709
Hopefully, it will resolve regressions since 4.8.1 (see also https://github.com/opencv/opencv/pull/24587)
### 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
dnn: add attention layer #24476Resolves#24609
Merge with: https://github.com/opencv/opencv_extra/pull/1128.
Attention operator spec from onnxruntime: https://github.com/microsoft/onnxruntime/blob/v1.16.1/docs/ContribOperators.md#com.microsoft.Attention.
TODO:
- [x] benchmark (before this PR vs. with this PR vs. ORT).
- [x] Layer fusion: Take care Slice with end=INT64_MAX.
- [x] Layer fusion: match more potential attention (VIT) patterns.
- [x] Single-head attention is supported.
- [x] Test AttentionSubgraph fusion.
- [x] Add acc tests for VIT_B_32 and VitTrack
- [x] Add perf tests for VIT_B_32 and VitTrack
## Benchmarks
Platform: Macbook Air M1.
### Attention Subgraph
Input scale: [1, 197, 768].
| | mean (ms) | median (ms) | min (ms) |
| ---------------------- | --------- | ----------- | -------- |
| w/ Attention (this PR) | 3.75 | 3.68 | 3.22 |
| w/o Attention | 9.06 | 9.01 | 8.24 |
| ORT (python) | 4.32 | 2.63 | 2.50 |
### ViTs
All data in millisecond (ms).
| ViTs | With Attention | Without Attention | ORT |
| -------- | -------------- | ----------------- | ------ |
| vit_b_16 | 302.77 | 365.35 | 109.70 |
| vit_b_32 | 89.92 | 116.22 | 30.36 |
| vit_l_16 | 1593.32 | 1730.74 | 419.92 |
| vit_l_32 | 468.11 | 577.41 | 134.12 |
| VitTrack | 3.80 | 3.87 | 2.25 |
### 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
dnn: refactor ONNX MatMul with fastGemm #24694
Done:
- [x] add backends
- [x] CUDA
- [x] OpenVINO
- [x] CANN
- [x] OpenCL
- [x] Vulkan
- [x] add perf tests
- [x] const B case
### Benchmark
Tests are done on M1. All data is in milliseconds (ms).
| Configuration | MatMul (Prepacked) | MatMul | InnerProduct |
| - | - | - | - |
| A=[12, 197, 197], B=[12, 197, 64], trans_a=0, trans_b=0 | **0.39** | 0.41 | 1.33 |
| A=[12, 197, 64], B=[12, 64, 197], trans_a=0, trans_b=0 | **0.42** | 0.42 | 1.17 |
| A=[12, 50, 64], B=[12, 64, 50], trans_a=0, trans_b=0 | **0.13** | 0.15 | 0.33 |
| A=[12, 50, 50], B=[12, 50, 64], trans_a=0, trans_b=0 | **0.11** | 0.13 | 0.22 |
| A=[16, 197, 197], B=[16, 197, 64], trans_a=0, trans_b=0 | **0.46** | 0.54 | 1.46 |
| A=[16, 197, 64], B=[16, 64, 197], trans_a=0, trans_b=0 | **0.46** | 0.95 | 1.74 |
| A=[16, 50, 64], B=[16, 64, 50], trans_a=0, trans_b=0 | **0.18** | 0.32 | 0.43 |
| A=[16, 50, 50], B=[16, 50, 64], trans_a=0, trans_b=0 | **0.15** | 0.25 | 0.25 |
### 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
* add Winograd FP16 implementation
* fixed dispatching of FP16 code paths in dnn; use dynamic dispatcher only when NEON_FP16 is enabled in the build and the feature is present in the host CPU at runtime
* fixed some warnings
* hopefully fixed winograd on x64 (and maybe other platforms)
---------
Co-authored-by: Vadim Pisarevsky <vadim.pisarevsky@gmail.com>
dnn onnx: add instance norm layer #24378
Resolves https://github.com/opencv/opencv/issues/24377
Relates https://github.com/opencv/opencv/pull/24092#discussion_r1349841644
| Perf | multi-thread | single-thread |
| - | - | - |
| x: [2, 64, 180, 240] | 3.95ms | 11.12ms |
Todo:
- [x] speed up by multi-threading
- [x] add perf
- [x] add backend: OpenVINO
- [x] add backend: CUDA
- [x] add backend: OpenCL (no fp16)
- [ ] add backend: CANN (will be done via https://github.com/opencv/opencv/pull/24462)
### 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
```
force_builders=Linux OpenCL,Win64 OpenCL,Custom
buildworker:Custom=linux-4
build_image:Custom=ubuntu:18.04
modules_filter:Custom=none
disable_ipp:Custom=ON
```
dnn: add shared fastNorm kernel for mvn, instance norm and layer norm #24409
Relates https://github.com/opencv/opencv/pull/24378#issuecomment-1756906570
TODO:
- [x] add fastNorm
- [x] refactor layer norm with fastNorm
- [x] refactor mvn with fastNorm
- [ ] add onnx mvn in importer (in a new PR?)
- [ ] refactor instance norm with fastNorm (in another PR https://github.com/opencv/opencv/pull/24378, need to merge this one first though)
### 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
GSoC Add ONNX Support for GatherElements #24092
Merge with: https://github.com/opencv/opencv_extra/pull/1082
Adds support to the ONNX operator GatherElements [operator docs](https://github.com/onnx/onnx/blob/main/docs/Operators.md#GatherElements)
Added tests to opencv_extra at pull request https://github.com/opencv/opencv_extra/pull/1082
### 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
Add Support for Einsum Layer #24037
### This PR adding support for [Einsum Layer](https://pytorch.org/docs/stable/generated/torch.einsum.html) (in progress).
This PR is currently not to be merged but only reviewed. Test cases are located in [#1090](https://github.com/opencv/opencv_extra/pull/1090)RP in OpenCV extra
**DONE**:
- [x] 2-5D GMM support added
- [x] Matrix transpose support added
- [x] Reduction type comupte 'ij->j'
- [x] 2nd shape computation - during forward
**Next PRs**:
- [ ] Broadcasting reduction "...ii ->...i"
- [ ] Add lazy shape deduction. "...ij, ...jk->...ik"
- [ ] Add implicit output computation support. "bij,bjk ->" (output subscripts should be "bik")
- [ ] Add support for CUDA backend
- [ ] BatchWiseMultiply optimize
**Later in 5.x version (requires support for 1D matrices)**:
- [ ] Add 1D vector multiplication support
- [ ] Inter product "i, i" (problems with 1D shapes)
### 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.
- [x] The feature is well documented and sample code can be built with the project CMake
* 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
dnn: add layer normalization for vision transformers
* add layer norm onnx parser, impl and tests
* add onnx graph simplifier for layer norm expanded
* handle the case when constants are of type Initializer
* add test case for layer norm expanded with initializers
* use CV_Assert & CV_CheckType in place of CV_Assert_N; use forward_fallback for OCL_FP16
* use const ref / ref in parameters of invoker::run; extract inner const if from nested loop; use size_t in place of ull
* template hasBias
* remove trailing whitespace
* use pointer parameter with null check; move normSize division & mean_square division outside of loop; use std::max to ensure positive value before std::sqrt
* refactor implementation, optimize parallel_for
* disable layer norm expanded
* remove the removal of layer norm optional outputs
DNN: let Quant and Dequant of ONNX_importer support the Constant input.
* let Quant and Dequant support the Constant input.
* fix negative value of axis.
Reimplementation of Element-wise layers with broadcasting support
* init
* semi-working initial version
* add small_vector
* wip
* remove smallvec
* add nary function
* replace auto with Mat in lambda expr used in transform
* uncomment asserts
* autobuffer shape_buf & step_buf
* fix a missing bracket
* fixed a missing addLayer in parseElementWise
* solve one-dimensional broadcast
* remove pre_broadcast_transform for the case of two constants; fix missing constBlobsExtraInfo when addConstant is called
* one autobuffer for step & shape
* temporal fix for the missing original dimension information
* fix parseUnsqueeze when it gets a 1d tensor constant
* support sum/mean/min/max with only one input
* reuse old code to handle cases of two non-constant inputs
* add condition to handle div & mul of two non-constant inputs
* use || instead of or
* remove trainling spaces
* enlarge buf in binary_forward to contain other buffer
* use autobuffer in nary_forward
* generate data randomly and add more cases for perf
* add op and, or & xor
* update perf_dnn
* remove some comments
* remove legacy; add two ONNX conformance tests in filter
* move from cpu_denylist to all_denylist
* adjust parsing for inputs>=2
Co-authored-by: fengyuentau <yuantao.feng@opencv.org.cn>
Add per_tensor_quantize to int8 quantize
* add per_tensor_quantize to dnn int8 module.
* change api flag from perTensor to perChannel, and recognize quantize type and onnx importer.
* change the default to hpp
dnn : int8 quantized layers support in onnx importer
* added quantized layers support in onnx importer
* added more cases in eltwise node, some more checks
* added tests for quantized nodes
* relax thresholds for failed tests, address review comments
* refactoring based on review comments
* added support for unsupported cases and pre-quantized resnet50 test
* relax thresholds due to int8 resize layer
Support for Pool1d layer for OpenCV and OpenCL targets
* Initial version of Pool1d support
* Fix variable naming
* Fix 1d pooling for OpenCL
* Change support logic, remove unnecessary variable, split the tests
* Remove other depricated variables
* Fix warning. Check tests
* Change support check logic
* Change support check logic, 2
