summary |
shortlog | log |
commit |
commitdiff |
tree
first ⋅ prev ⋅ next
Inki Dae [Thu, 4 Jun 2020 05:36:41 +0000 (14:36 +0900)]
Fix coding style based on Tizen SE C++ Coding Rule
Tizen SE C++ Coding Rule:
https://code.sec.samsung.net/confluence/pages/viewpage.action?pageId=
160925159
Change-Id: Ia01c4ef55138bdbccb8fdaebf5b3d86494bde001
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Inki Dae [Wed, 3 Jun 2020 08:55:55 +0000 (17:55 +0900)]
Add SetPrivateData function
This function is needed for inference engine interface to pass
opencv specific data to this backend before loading a model file.
Change-Id: Id95fe872d810929d41a02c4c30c494040f8fffe1
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Inki Dae [Fri, 24 Apr 2020 03:12:55 +0000 (12:12 +0900)]
fix build error on aarch64
In aarch64, size_t is 64bit data so make sure to use correct
format specifier.
Change-Id: I4990e9cd8556208f626707905cd423e81d23f65b
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Inki Dae [Fri, 17 Apr 2020 07:43:07 +0000 (16:43 +0900)]
change enumeration values to new ones
some enumeration values of inference-engine-interface have been
updated so change them.
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Inki Dae [Fri, 17 Apr 2020 07:03:55 +0000 (16:03 +0900)]
Fix build error
shape type of inference_engine_tensor_info has been changed
from vector<int> to vector<size_t> so correct it.
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Tae-Young Chung [Wed, 8 Apr 2020 04:17:10 +0000 (13:17 +0900)]
Add dummy buffer to inference_engine_tensor_buffers for output buffers
In case of object detection, mobilenet-ssd, its output may be null and
it causes the memory allocation at inference-engine-interface
not from opencv itself.
To handle that case, temporally add dummy buffer.
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tae-Young Chung [Thu, 2 Apr 2020 05:38:47 +0000 (14:38 +0900)]
Fixed wrong input/output buffer allocation
Fixed wrong input/output buffer size in GetInputTensorBuffers() and
GetOutputTensorBuffers(). In Run(), the output buffer addr is changed
after forward() so buffer addr should be updated.
Add a trick to handle output without post-process.
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tae-Young Chung [Wed, 4 Mar 2020 05:49:12 +0000 (14:49 +0900)]
Refactoring code according to refactored inferenc-engine-interface
NOTE that OpenCV doesn't support to get information of input layer.
The input should be set by calling cv::dnn::Net::setInput() API.
OpenCV allocates memory on the first cv::dnn::Net::forward().
After that, it reuse the allocated memory.
So, in this refactoring,
Create mInputData as cv::Mat type for input tensors
and returns their allocated memories.
Allocate the memories of mOutputBlobs by calling forward() with
a dummy tensors.
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tae-Young Chung [Tue, 7 Apr 2020 04:00:46 +0000 (13:00 +0900)]
Keep opencv.pc file
Change-Id: Ib150d486e6931c8407d662daa1ceac7b024fc0a7
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tae-Young Chung [Thu, 2 Jan 2020 08:34:49 +0000 (17:34 +0900)]
Migration with OpenCV-4.2.0
Change-Id: I4a703b39f7328dea2facc8df7e5f8c5c29f71f99
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tae-Young Chung [Mon, 23 Sep 2019 06:07:55 +0000 (15:07 +0900)]
Remove apis related to IInferenceEngineVision
inference-engine-opencv is a plugin to provide inference only.
Thus, domain specific functions such as vision should be removed
even though it is OpenCV.
Instead, add apis GetInputLayerAttrType(), SetInputDataBuffer(), GetInputDataPtr()
which can be used to access to memory.
Change-Id: I8abaa91b10396a2687b25e10d50a31d70405c2ed
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Hyunsoo Park [Fri, 6 Sep 2019 06:28:24 +0000 (15:28 +0900)]
Sets NULL to pointer variable.
C++ delete method just release allocated memory. It doesn't set NULL value to variable.
So i make variable to NULL after delete process.
Change-Id: I42661f77ff3bcc2b0dafbb3b37741cc8c0d4e720
Signed-off-by: Hyunsoo Park <hance.park@samsung.com>
Hyunsoo Park [Thu, 5 Sep 2019 05:06:23 +0000 (14:06 +0900)]
Remove PrepareInputLayerPassage().
It doesn't use anymore. So i remove it.
Change-Id: If6113a2982e2aa0c3111cbedb5dcfd726128d1f0
Signed-off-by: Hyunsoo Park <hance.park@samsung.com>
Hyunsoo Park [Thu, 22 Aug 2019 11:33:07 +0000 (20:33 +0900)]
Deletion of related methods as inherited class changing.
It is inherited by IInferenceEngineVision and IInferenceEngineCommon both.
But relationship of them is changed as inheritant. So i remove unneccessary methods.
Change-Id: I0821a4d3d7b0640cabdeadbef180fc6fd097f90d
Signed-off-by: Hyunsoo Park <hance.park@samsung.com>
Tae-Young Chung [Thu, 25 Jul 2019 05:54:11 +0000 (14:54 +0900)]
Add INFERENCE_TARGET_CUSTOM, Support multiple output tensors
Change-Id: Iab1965c027b4207ae699f5998f789b85020f7a8b
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tae-Young Chung [Thu, 4 Jul 2019 11:15:34 +0000 (20:15 +0900)]
Initial codes
Note that this initial code is under development and provides limited functionality.
In addition, OpenCV based implementation must be applied to vision specific purpose.
Signed-off-by: Tae-Young Chung <ty83.chung@samsung.com>
Tizen Infrastructure [Mon, 1 Jul 2019 01:54:16 +0000 (01:54 +0000)]
Initial empty repository