Merge pull request #18952 from terfendail:wui_doc

* Updated UI documentation to address WUI

* Added documentation for vx_ calls

* Removed vx_store operation overload

* Doxyfile updated to enable wide UI

* Enable doxygen documentation for vx_ WUI functions

* Wide intrinsics definition rework

* core: fix SIMD C++ emulator build (supports 128-bit only)

doc/Doxyfile.in

@@ -249,6 +249,12 @@ PREDEFINED             = __cplusplus=1 \
                          CV_DEFAULT(x)=" = x" \
                          CV_NEON=1 \
                          CV_SSE2=1 \
+                         CV_SIMD128=1 \
+                         CV_SIMD256=1 \
+                         CV_SIMD512=1 \
+                         CV_SIMD128_64F=1 \
+                         CV_SIMD256_64F=1 \
+                         CV_SIMD512_64F=1 \
                          CV__DEBUG_NS_BEGIN= \
                          CV__DEBUG_NS_END= \
                          CV_DEPRECATED_EXTERNAL= \

modules/core/src/matrix_transform.cpp

@@ -536,8 +536,8 @@ flipVert( const uchar* src0, size_t sstep, uchar* dst0, size_t dstep, Size size,
             {
                 v_int32 t0 = vx_load((int*)(src0 + i));
                 v_int32 t1 = vx_load((int*)(src1 + i));
-                vx_store((int*)(dst0 + i), t1);
-                vx_store((int*)(dst1 + i), t0);
+                v_store((int*)(dst0 + i), t1);
+                v_store((int*)(dst1 + i), t0);
             }
         }
 #if CV_STRONG_ALIGNMENT
@@ -547,8 +547,8 @@ flipVert( const uchar* src0, size_t sstep, uchar* dst0, size_t dstep, Size size,
             {
                 v_uint8 t0 = vx_load(src0 + i);
                 v_uint8 t1 = vx_load(src1 + i);
-                vx_store(dst0 + i, t1);
-                vx_store(dst1 + i, t0);
+                v_store(dst0 + i, t1);
+                v_store(dst1 + i, t0);
             }
         }
 #endif

modules/core/test/test_intrin_utils.hpp

@@ -1466,7 +1466,7 @@ template<typename R> struct TheTest
         R r1 = vx_load_expand((const cv::float16_t*)data.a.d);
         R r2(r1);
         EXPECT_EQ(1.0f, r1.get0());
-        vx_store(data_f32.a.d, r2);
+        v_store(data_f32.a.d, r2);
         EXPECT_EQ(-2.0f, data_f32.a.d[R::nlanes - 1]);
 
         out.a.clear();

modules/dnn/src/layers/convolution_layer.cpp

@@ -1023,7 +1023,7 @@ public:
                                                              v20*vw20 + v21*vw21 + v22*vw22 + vbias;
                                             if (relu)
                                                 vout = v_select(vout > z, vout, vout*vrc);
-                                            vx_store(outptr + out_j, vout);
+                                            v_store(outptr + out_j, vout);
                                         }
                                     }
                                 #endif