Merge remote-tracking branch 'upstream/3.4' into merge-3.4

advait-0 · Dec 15, 2021 · 299f983 · 299f983
2 parents 498f152 + 6a889ed
commit 299f983
Show file tree

Hide file tree

Showing 26 changed files with 4,405 additions and 18 deletions.
diff --git a/cmake/OpenCVDetectInferenceEngine.cmake b/cmake/OpenCVDetectInferenceEngine.cmake
@@ -106,8 +106,8 @@ if(DEFINED InferenceEngine_VERSION)
   endif()
 endif()
 if(NOT INF_ENGINE_RELEASE AND NOT INF_ENGINE_RELEASE_INIT)
-  message(STATUS "WARNING: InferenceEngine version has not been set, 2021.4.1 will be used by default. Set INF_ENGINE_RELEASE variable if you experience build errors.")
-  set(INF_ENGINE_RELEASE_INIT "2021040100")
+  message(STATUS "WARNING: InferenceEngine version has not been set, 2021.4.2 will be used by default. Set INF_ENGINE_RELEASE variable if you experience build errors.")
+  set(INF_ENGINE_RELEASE_INIT "2021040200")
 elseif(DEFINED INF_ENGINE_RELEASE)
   set(INF_ENGINE_RELEASE_INIT "${INF_ENGINE_RELEASE}")
 endif()

diff --git a/doc/tutorials/imgproc/pyramids/pyramids.markdown b/doc/tutorials/imgproc/pyramids/pyramids.markdown
@@ -54,7 +54,7 @@ Theory
 -   To produce layer \f$(i+1)\f$ in the Gaussian pyramid, we do the following:
     -   Convolve \f$G_{i}\f$ with a Gaussian kernel:
 
-        \f[\frac{1}{16} \begin{bmatrix} 1 & 4 & 6 & 4 & 1  \\ 4 & 16 & 24 & 16 & 4  \\ 6 & 24 & 36 & 24 & 6  \\ 4 & 16 & 24 & 16 & 4  \\ 1 & 4 & 6 & 4 & 1 \end{bmatrix}\f]
+        \f[\frac{1}{256} \begin{bmatrix} 1 & 4 & 6 & 4 & 1  \\ 4 & 16 & 24 & 16 & 4  \\ 6 & 24 & 36 & 24 & 6  \\ 4 & 16 & 24 & 16 & 4  \\ 1 & 4 & 6 & 4 & 1 \end{bmatrix}\f]
 
     -   Remove every even-numbered row and column.
 

diff --git a/modules/dnn/src/layers/padding_layer.cpp b/modules/dnn/src/layers/padding_layer.cpp
@@ -140,12 +140,13 @@ class PaddingLayerImpl CV_FINAL : public PaddingLayer
                 outputs[0].setTo(paddingValue);
             inputs[0].copyTo(outputs[0](dstRanges));
         }
-        else if (paddingType == "reflect")
+        else if (paddingType == "reflect" || paddingType == "edge")
         {
             CV_Assert(inputs.size() == 1);
             CV_Assert(outputs.size() == 1);
             CV_Assert(inputs[0].dims == 4);
             CV_Assert(outputs[0].dims == 4);
+            int borderType = paddingType == "reflect" ? BORDER_REFLECT_101 : BORDER_REPLICATE;
 
             if (inputs[0].size[0] != outputs[0].size[0] || inputs[0].size[1] != outputs[0].size[1])
                 CV_Error(Error::StsNotImplemented, "Only spatial reflection padding is supported.");
@@ -158,8 +159,8 @@ class PaddingLayerImpl CV_FINAL : public PaddingLayer
             const int padBottom = outHeight - dstRanges[2].end;
             const int padLeft = dstRanges[3].start;
             const int padRight = outWidth - dstRanges[3].end;
-            CV_CheckLT(padTop, inpHeight, ""); CV_CheckLT(padBottom, inpHeight, "");
-            CV_CheckLT(padLeft, inpWidth, ""); CV_CheckLT(padRight, inpWidth, "");
+            CV_CheckLE(padTop, inpHeight, ""); CV_CheckLE(padBottom, inpHeight, "");
+            CV_CheckLE(padLeft, inpWidth, ""); CV_CheckLE(padRight, inpWidth, "");
 
             for (size_t n = 0; n < inputs[0].size[0]; ++n)
             {
@@ -168,7 +169,7 @@ class PaddingLayerImpl CV_FINAL : public PaddingLayer
                     copyMakeBorder(getPlane(inputs[0], n, ch),
                                    getPlane(outputs[0], n, ch),
                                    padTop, padBottom, padLeft, padRight,
-                                   BORDER_REFLECT_101);
+                                   borderType);
                 }
             }
         }

diff --git a/modules/dnn/src/onnx/onnx_importer.cpp b/modules/dnn/src/onnx/onnx_importer.cpp
@@ -101,6 +101,7 @@ class ONNXImporter
     static const DispatchMap buildDispatchMap();
 
     void parseArg                  (LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto);
+    void parseMaxUnpool            (LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto);
     void parseMaxPool              (LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto);
     void parseAveragePool          (LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto);
     void parseReduce               (LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto);
@@ -790,6 +791,41 @@ void setCeilMode(LayerParams& layerParams)
     }
 }
 
+void ONNXImporter::parseMaxUnpool(LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto)
+{
+    layerParams.type = "MaxUnpool";
+
+    DictValue kernel_shape = layerParams.get("kernel_size");
+    CV_Assert(kernel_shape.size() == 2);
+    layerParams.set("pool_k_w", kernel_shape.get<int>(0));
+    layerParams.set("pool_k_h", kernel_shape.get<int>(1));
+
+    int pool_pad_w = 0, pool_pad_h = 0;
+    if (layerParams.has("pad"))
+    {
+        DictValue pads = layerParams.get("pad");
+        CV_CheckEQ(pads.size(), 2, "");
+        pool_pad_w = pads.get<int>(0);
+        pool_pad_h = pads.get<int>(1);
+    }
+    layerParams.set("pool_pad_w", pool_pad_w);
+    layerParams.set("pool_pad_h", pool_pad_h);
+
+
+    int pool_stride_w = 1, pool_stride_h = 1;
+    if (layerParams.has("stride"))
+    {
+        DictValue strides = layerParams.get("stride");
+        CV_CheckEQ(strides.size(), 2, "");
+        pool_stride_w = strides.get<int>(0);
+        pool_stride_h = strides.get<int>(1);
+    }
+    layerParams.set("pool_stride_w", pool_stride_w);
+    layerParams.set("pool_stride_h", pool_stride_h);
+
+    addLayer(layerParams, node_proto);
+}
+
 void ONNXImporter::parseMaxPool(LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto)
 {
     int depth = layerParams.get<int>("depth", CV_32F);
@@ -824,11 +860,11 @@ void ONNXImporter::parseReduce(LayerParams& layerParams, const opencv_onnx::Node
         pool = "AVE";
     layerParams.set("pool", pool);
     layerParams.set("global_pooling", !layerParams.has("axes"));
+    bool keepdims = layerParams.get<int>("keepdims", 1) == 1;
     if (layerParams.has("axes") && (layer_type == "ReduceMean" || layer_type == "ReduceSum" || layer_type == "ReduceMax"))
     {
         MatShape inpShape = outShapes[node_proto.input(0)];
         DictValue axes = layerParams.get("axes");
-        bool keepdims = layerParams.get<int>("keepdims");
         MatShape targetShape;
         std::vector<bool> shouldDelete(inpShape.size(), false);
         for (int i = 0; i < axes.size(); i++) {
@@ -936,7 +972,10 @@ void ONNXImporter::parseReduce(LayerParams& layerParams, const opencv_onnx::Node
     }
     else if (!layerParams.has("axes") && (layer_type == "ReduceMean" || layer_type == "ReduceSum" || layer_type == "ReduceMax"))
     {
-        CV_CheckEQ(layerParams.get<int>("keepdims"), 0, "layer only supports keepdims = false");
+        IterShape_t shapeIt = outShapes.find(node_proto.input(0));
+        CV_Assert(shapeIt != outShapes.end());
+        const size_t dims = keepdims ? shapeIt->second.size() : 1;
+
         LayerParams reshapeLp;
         reshapeLp.name = layerParams.name + "/reshape";
         reshapeLp.type = "Reshape";
@@ -958,8 +997,8 @@ void ONNXImporter::parseReduce(LayerParams& layerParams, const opencv_onnx::Node
         addLayer(poolLp, node_proto);
 
         layerParams.type = "Reshape";
-        int targetShape[] = {1};
-        layerParams.set("dim", DictValue::arrayInt(&targetShape[0], 1));
+        std::vector<int> targetShape(dims, 1);
+        layerParams.set("dim", DictValue::arrayInt(targetShape.data(), targetShape.size()));
 
         node_proto.set_input(0, node_proto.output(0));
         node_proto.set_output(0, layerParams.name);
@@ -1412,6 +1451,7 @@ void ONNXImporter::parseImageScaler(LayerParams& layerParams, const opencv_onnx:
 
 void ONNXImporter::parseClip(LayerParams& layerParams, const opencv_onnx::NodeProto& node_proto)
 {
+    CV_CheckEQ(node_proto.input_size(), 1, "");
     layerParams.type = "ReLU6";
     layerParams.set("min_value", layerParams.get<float>("min", -FLT_MAX));
     layerParams.set("max_value", layerParams.get<float>("max", FLT_MAX));
@@ -2996,6 +3036,7 @@ const ONNXImporter::DispatchMap ONNXImporter::buildDispatchMap()
     DispatchMap dispatch;
 
     dispatch["ArgMax"] = dispatch["ArgMin"] = &ONNXImporter::parseArg;
+    dispatch["MaxUnpool"] = &ONNXImporter::parseMaxUnpool;
     dispatch["MaxPool"] = &ONNXImporter::parseMaxPool;
     dispatch["AveragePool"] = &ONNXImporter::parseAveragePool;
     dispatch["GlobalAveragePool"] = dispatch["GlobalMaxPool"] = dispatch["ReduceMean"] = dispatch["ReduceSum"] =

diff --git a/modules/dnn/test/test_common.hpp b/modules/dnn/test/test_common.hpp
@@ -18,8 +18,9 @@
 #define INF_ENGINE_VER_MAJOR_LE(ver) (((INF_ENGINE_RELEASE) / 10000) <= ((ver) / 10000))
 #define INF_ENGINE_VER_MAJOR_EQ(ver) (((INF_ENGINE_RELEASE) / 10000) == ((ver) / 10000))
 
-
+#define CV_TEST_TAG_DNN_SKIP_OPENCV_BACKEND      "dnn_skip_opencv_backend"
 #define CV_TEST_TAG_DNN_SKIP_HALIDE              "dnn_skip_halide"
+#define CV_TEST_TAG_DNN_SKIP_CPU                 "dnn_skip_cpu"
 #define CV_TEST_TAG_DNN_SKIP_OPENCL              "dnn_skip_ocl"
 #define CV_TEST_TAG_DNN_SKIP_OPENCL_FP16         "dnn_skip_ocl_fp16"
 #define CV_TEST_TAG_DNN_SKIP_IE_NN_BUILDER       "dnn_skip_ie_nn_builder"
@@ -44,6 +45,9 @@
 #define CV_TEST_TAG_DNN_SKIP_CUDA_FP16           "dnn_skip_cuda_fp16"
 #define CV_TEST_TAG_DNN_SKIP_CUDA_FP32           "dnn_skip_cuda_fp32"
 
+#define CV_TEST_TAG_DNN_SKIP_ONNX_CONFORMANCE    "dnn_skip_onnx_conformance"
+#define CV_TEST_TAG_DNN_SKIP_PARSER              "dnn_skip_parser"
+
 
 #ifdef HAVE_INF_ENGINE
 #if INF_ENGINE_VER_MAJOR_EQ(2018050000)

diff --git a/modules/dnn/test/test_common.impl.hpp b/modules/dnn/test/test_common.impl.hpp
@@ -442,9 +442,15 @@ void initDNNTests()
         cvtest::addDataSearchPath(extraTestDataPath);
 
     registerGlobalSkipTag(
-        CV_TEST_TAG_DNN_SKIP_HALIDE,
+        CV_TEST_TAG_DNN_SKIP_OPENCV_BACKEND,
+        CV_TEST_TAG_DNN_SKIP_CPU,
         CV_TEST_TAG_DNN_SKIP_OPENCL, CV_TEST_TAG_DNN_SKIP_OPENCL_FP16
     );
+#if defined(HAVE_HALIDE)
+    registerGlobalSkipTag(
+        CV_TEST_TAG_DNN_SKIP_HALIDE
+    );
+#endif
 #if defined(INF_ENGINE_RELEASE)
     registerGlobalSkipTag(
         CV_TEST_TAG_DNN_SKIP_IE,
@@ -478,12 +484,15 @@ void initDNNTests()
         CV_TEST_TAG_DNN_SKIP_VULKAN
     );
 #endif
-
 #ifdef HAVE_CUDA
     registerGlobalSkipTag(
         CV_TEST_TAG_DNN_SKIP_CUDA, CV_TEST_TAG_DNN_SKIP_CUDA_FP32, CV_TEST_TAG_DNN_SKIP_CUDA_FP16
     );
 #endif
+    registerGlobalSkipTag(
+        CV_TEST_TAG_DNN_SKIP_ONNX_CONFORMANCE,
+        CV_TEST_TAG_DNN_SKIP_PARSER
+    );
 }
 
 } // namespace