diff --git a/include/sparkyuv-basic.h b/include/sparkyuv-basic.h
index ec16853..e6d3a52 100644
--- a/include/sparkyuv-basic.h
+++ b/include/sparkyuv-basic.h
@@ -151,6 +151,38 @@ FLIP_HORIZONTAL_DECLARATION_H(RGB16, uint16_t, CHANNELS_3)
 FLIP_HORIZONTAL_DECLARATION_H(Channel16, uint16_t, CHANNEL)
 
 #undef FLIP_HORIZONTAL_DECLARATION_H
+
+#define TRANSPOSE_CLOCKWISE_DECLARATION_H(srcPixel, storageType) \
+    void TransposeClockwise##srcPixel(const storageType* src, const uint32_t srcStride,\
+                                      storageType * dst, const uint32_t dstStride,\
+                                      const uint32_t width, const uint32_t height);
+
+TRANSPOSE_CLOCKWISE_DECLARATION_H(RGBA, uint8_t)
+TRANSPOSE_CLOCKWISE_DECLARATION_H(RGB, uint8_t)
+TRANSPOSE_CLOCKWISE_DECLARATION_H(Channel, uint8_t)
+TRANSPOSE_CLOCKWISE_DECLARATION_H(RGBA1010102, uint8_t)
+
+TRANSPOSE_CLOCKWISE_DECLARATION_H(RGBA16, uint16_t)
+TRANSPOSE_CLOCKWISE_DECLARATION_H(RGB16, uint16_t)
+TRANSPOSE_CLOCKWISE_DECLARATION_H(Channel16, uint16_t)
+
+#undef TRANSPOSE_CLOCKWISE_DECLARATION_H
+
+#define TRANSPOSE_CCW_DECLARATION_H(srcPixel, storageType) \
+    void TransposeCounterClockwise##srcPixel(const storageType * src, const uint32_t srcStride,\
+                                           storageType * dst, const uint32_t dstStride,\
+                                           const uint32_t width, const uint32_t height);
+
+TRANSPOSE_CCW_DECLARATION_H(RGBA, uint8_t)
+TRANSPOSE_CCW_DECLARATION_H(RGB, uint8_t)
+TRANSPOSE_CCW_DECLARATION_H(Channel, uint8_t)
+TRANSPOSE_CCW_DECLARATION_H(RGBA1010102, uint8_t)
+
+TRANSPOSE_CCW_DECLARATION_H(RGBA16, uint16_t)
+TRANSPOSE_CCW_DECLARATION_H(RGB16, uint16_t)
+TRANSPOSE_CCW_DECLARATION_H(Channel16, uint16_t)
+
+#undef TRANSPOSE_CCW_DECLARATION_H
 }
 
 #endif //YUV_INCLUDE_SPARKYUV_BASIC_H_
diff --git a/src/Transpose-inl.h b/src/Transpose-inl.h
index 2a6429a..e769912 100644
--- a/src/Transpose-inl.h
+++ b/src/Transpose-inl.h
@@ -28,6 +28,122 @@
 
 HWY_BEFORE_NAMESPACE();
 namespace sparkyuv::HWY_NAMESPACE {
+using namespace hwy;
+using namespace hwy::HWY_NAMESPACE;
+
+template<class T, SparkYuvSurfaceChannels Surface = sparkyuv::SURFACE_CHANNEL>
+void
+TransposeClockwiseImpl(const T *SPARKYUV_RESTRICT src, const uint32_t srcStride,
+                       T *SPARKYUV_RESTRICT dst, const uint32_t newStride,
+                       const uint32_t width, const uint32_t height) {
+
+  auto mSourcePtr = reinterpret_cast<const uint8_t *>(src);
+  auto mDestinationPtr = reinterpret_cast<uint8_t *>(dst);
+
+  const int channels = Surface == SURFACE_CHANNEL ? 1 : (Surface == SURFACE_CHANNELS_3 ? 3 : 4);
+
+  for (uint32_t y = 0; y < height; ++y) {
+    auto mSrc = reinterpret_cast<const T *>(src + y * srcStride);
+    for (uint32_t x = 0; x < width; ++x) {
+
+      auto mTransposed = reinterpret_cast<T *>(mDestinationPtr + x * newStride);
+      mTransposed += (height - 1 - y) * channels;
+
+      if (Surface == SURFACE_CHANNEL) {
+        mTransposed[0] = mSrc[0];
+      } else if (Surface == SURFACE_CHANNELS_3) {
+        mTransposed[0] = mSrc[0];
+        mTransposed[1] = mSrc[1];
+        mTransposed[2] = mSrc[2];
+      } else if (Surface == SURFACE_CHANNELS_4) {
+        mTransposed[0] = mSrc[0];
+        mTransposed[1] = mSrc[1];
+        mTransposed[2] = mSrc[2];
+        mTransposed[3] = mSrc[3];
+      }
+
+      mSrc += channels;
+    }
+
+    mSourcePtr += srcStride;
+  }
+}
+
+#define TRANSPOSE_CLOCKWISE_DECLARATION(srcPixel, storageType, surfaceType) \
+    void TransposeClockwise##srcPixel##HWY(const storageType *SPARKYUV_RESTRICT src, const uint32_t srcStride,\
+                                           storageType *SPARKYUV_RESTRICT dst, const uint32_t dstStride,\
+                                           const uint32_t width, const uint32_t height) {\
+        TransposeClockwiseImpl<storageType, sparkyuv::SURFACE_##surfaceType>(src, srcStride, dst, dstStride,\
+                                                                             width, height); \
+    }
+
+TRANSPOSE_CLOCKWISE_DECLARATION(RGBA, uint8_t, CHANNELS_4)
+TRANSPOSE_CLOCKWISE_DECLARATION(RGB, uint8_t, CHANNELS_3)
+TRANSPOSE_CLOCKWISE_DECLARATION(Channel, uint8_t, CHANNEL)
+TRANSPOSE_CLOCKWISE_DECLARATION(RGBA1010102, uint8_t, RGBA1010102)
+
+TRANSPOSE_CLOCKWISE_DECLARATION(RGBA16, uint16_t, CHANNELS_4)
+TRANSPOSE_CLOCKWISE_DECLARATION(RGB16, uint16_t, CHANNELS_3)
+TRANSPOSE_CLOCKWISE_DECLARATION(Channel16, uint16_t, CHANNEL)
+
+#undef TRANSPOSE_CLOCKWISE_DECLARATION
+
+template<class T, SparkYuvSurfaceChannels Surface = sparkyuv::SURFACE_CHANNEL>
+void
+TransposeCounterClockwiseImpl(const T *SPARKYUV_RESTRICT src, const uint32_t srcStride,
+                              T *SPARKYUV_RESTRICT dst, const uint32_t newStride,
+                              const uint32_t width, const uint32_t height) {
+
+  auto mSourcePtr = reinterpret_cast<const uint8_t *>(src);
+  auto mDestinationPtr = reinterpret_cast<uint8_t *>(dst);
+
+  const int channels = Surface == SURFACE_CHANNEL ? 1 : (Surface == SURFACE_CHANNELS_3 ? 3 : 4);
+
+  for (uint32_t y = 0; y < height; ++y) {
+    auto mSrc = reinterpret_cast<const T *>(src + y * srcStride);
+    for (uint32_t x = 0; x < width; ++x) {
+
+      auto mTransposed = reinterpret_cast<T *>(mDestinationPtr + (width - 1 - x) * newStride);
+      mTransposed += y * channels;
+
+      if (Surface == SURFACE_CHANNEL) {
+        mTransposed[0] = mSrc[0];
+      } else if (Surface == SURFACE_CHANNELS_3) {
+        mTransposed[0] = mSrc[0];
+        mTransposed[1] = mSrc[1];
+        mTransposed[2] = mSrc[2];
+      } else if (Surface == SURFACE_CHANNELS_4) {
+        mTransposed[0] = mSrc[0];
+        mTransposed[1] = mSrc[1];
+        mTransposed[2] = mSrc[2];
+        mTransposed[3] = mSrc[3];
+      }
+
+      mSrc += channels;
+    }
+
+    mSourcePtr += srcStride;
+  }
+}
+
+#define TRANSPOSE_CCW_DECLARATION(srcPixel, storageType, surfaceType) \
+    void TransposeCounterClockwise##srcPixel##HWY(const storageType *SPARKYUV_RESTRICT src, const uint32_t srcStride,\
+                                           storageType *SPARKYUV_RESTRICT dst, const uint32_t dstStride,\
+                                           const uint32_t width, const uint32_t height) {\
+        TransposeCounterClockwiseImpl<storageType, sparkyuv::SURFACE_##surfaceType>(src, srcStride, dst, dstStride,\
+                                                                             width, height); \
+    }
+
+TRANSPOSE_CCW_DECLARATION(RGBA, uint8_t, CHANNELS_4)
+TRANSPOSE_CCW_DECLARATION(RGB, uint8_t, CHANNELS_3)
+TRANSPOSE_CCW_DECLARATION(Channel, uint8_t, CHANNEL)
+TRANSPOSE_CCW_DECLARATION(RGBA1010102, uint8_t, RGBA1010102)
+
+TRANSPOSE_CCW_DECLARATION(RGBA16, uint16_t, CHANNELS_4)
+TRANSPOSE_CCW_DECLARATION(RGB16, uint16_t, CHANNELS_3)
+TRANSPOSE_CCW_DECLARATION(Channel16, uint16_t, CHANNEL)
+
+#undef TRANSPOSE_CCW_DECLARATION
 
 }
 HWY_AFTER_NAMESPACE();
diff --git a/src/Transpose.cpp b/src/Transpose.cpp
index 9b73b72..c942385 100644
--- a/src/Transpose.cpp
+++ b/src/Transpose.cpp
@@ -25,6 +25,71 @@
 
 #if HWY_ONCE
 namespace sparkyuv {
+#define TRANSPOSE_CLOCKWISE_EXPORT(srcPixel, storageType)\
+  HWY_EXPORT(TransposeClockwise##srcPixel##HWY);
+
+TRANSPOSE_CLOCKWISE_EXPORT(RGBA, uint8_t)
+TRANSPOSE_CLOCKWISE_EXPORT(RGB, uint8_t)
+TRANSPOSE_CLOCKWISE_EXPORT(Channel, uint8_t)
+TRANSPOSE_CLOCKWISE_EXPORT(RGBA1010102, uint8_t)
+
+TRANSPOSE_CLOCKWISE_EXPORT(RGBA16, uint16_t)
+TRANSPOSE_CLOCKWISE_EXPORT(RGB16, uint16_t)
+TRANSPOSE_CLOCKWISE_EXPORT(Channel16, uint16_t)
+
+#undef TRANSPOSE_CLOCKWISE_EXPORT
+
+#define TRANSPOSE_CLOCKWISE_DECLARATION_R(srcPixel, storageType, surfaceType) \
+    void TransposeClockwise##srcPixel(const storageType *SPARKYUV_RESTRICT src, const uint32_t srcStride,\
+                                      storageType *SPARKYUV_RESTRICT dst, const uint32_t dstStride,\
+                                      const uint32_t width, const uint32_t height) {\
+        HWY_DYNAMIC_DISPATCH(TransposeClockwise##srcPixel##HWY)(src, srcStride, dst, dstStride,\
+                                                                width, height); \
+    }
+
+TRANSPOSE_CLOCKWISE_DECLARATION_R(RGBA, uint8_t, CHANNELS_4)
+TRANSPOSE_CLOCKWISE_DECLARATION_R(RGB, uint8_t, CHANNELS_3)
+TRANSPOSE_CLOCKWISE_DECLARATION_R(Channel, uint8_t, CHANNEL)
+TRANSPOSE_CLOCKWISE_DECLARATION_R(RGBA1010102, uint8_t, RGBA1010102)
+
+TRANSPOSE_CLOCKWISE_DECLARATION_R(RGBA16, uint16_t, CHANNELS_4)
+TRANSPOSE_CLOCKWISE_DECLARATION_R(RGB16, uint16_t, CHANNELS_3)
+TRANSPOSE_CLOCKWISE_DECLARATION_R(Channel16, uint16_t, CHANNEL)
+
+#undef TRANSPOSE_CLOCKWISE_DECLARATION_R
+
+#define TRANSPOSE_CCW_EXPORT(srcPixel, storageType)\
+  HWY_EXPORT(TransposeCounterClockwise##srcPixel##HWY);
+
+TRANSPOSE_CCW_EXPORT(RGBA, uint8_t)
+TRANSPOSE_CCW_EXPORT(RGB, uint8_t)
+TRANSPOSE_CCW_EXPORT(Channel, uint8_t)
+TRANSPOSE_CCW_EXPORT(RGBA1010102, uint8_t)
+
+TRANSPOSE_CCW_EXPORT(RGBA16, uint16_t)
+TRANSPOSE_CCW_EXPORT(RGB16, uint16_t)
+TRANSPOSE_CCW_EXPORT(Channel16, uint16_t)
+
+#undef TRANSPOSE_CCW_EXPORT
+
+#define TRANSPOSE_CCW_DECLARATION_E(srcPixel, storageType, surfaceType) \
+    void TransposeCounterClockwise##srcPixel(const storageType *SPARKYUV_RESTRICT src, const uint32_t srcStride,\
+                                           storageType *SPARKYUV_RESTRICT dst, const uint32_t dstStride,\
+                                           const uint32_t width, const uint32_t height) {\
+        HWY_DYNAMIC_DISPATCH(TransposeCounterClockwise##srcPixel##HWY)(src, srcStride, dst, dstStride,\
+                                                                       width, height); \
+    }
+
+TRANSPOSE_CCW_DECLARATION_E(RGBA, uint8_t, CHANNELS_4)
+TRANSPOSE_CCW_DECLARATION_E(RGB, uint8_t, CHANNELS_3)
+TRANSPOSE_CCW_DECLARATION_E(Channel, uint8_t, CHANNEL)
+TRANSPOSE_CCW_DECLARATION_E(RGBA1010102, uint8_t, RGBA1010102)
+
+TRANSPOSE_CCW_DECLARATION_E(RGBA16, uint16_t, CHANNELS_4)
+TRANSPOSE_CCW_DECLARATION_E(RGB16, uint16_t, CHANNELS_3)
+TRANSPOSE_CCW_DECLARATION_E(Channel16, uint16_t, CHANNEL)
+
+#undef TRANSPOSE_CCW_DECLARATION_E
 
 }
 #endif
\ No newline at end of file
diff --git a/tools/main.cpp b/tools/main.cpp
index 3224b5e..5df6935 100644
--- a/tools/main.cpp
+++ b/tools/main.cpp
@@ -151,9 +151,9 @@ int main() {
 
   bench(1, ANSI_COLOR_GREEN, "RGBA -> YCbCr420", [&]() {
     sparkyuv::RGBAToYCbCr444(rgbaData.data(), rgbaStride, width, height,
-                              yPlane.data(), yPlaneStride,
-                              uPlane.data(), uvPlaneStride,
-                              vPlane.data(), uvPlaneStride, 0.299f, 0.114f, sparkyuv::YUV_RANGE_TV);
+                             yPlane.data(), yPlaneStride,
+                             uPlane.data(), uvPlaneStride,
+                             vPlane.data(), uvPlaneStride, 0.299f, 0.114f, sparkyuv::YUV_RANGE_TV);
 //    libyuv::ABGRToI420(rgbaData.data(), rgbaStride, yPlane.data(), yPlaneStride,
 //                       uPlane.data(), uvPlaneStride,
 //                       vPlane.data(), uvPlaneStride, width, height );
@@ -166,9 +166,9 @@ int main() {
 //                       uPlane.data(), uvPlaneStride,
 //                       vPlane.data(), uvPlaneStride,  rgbaData.data(), rgbaStride, width, height);
     sparkyuv::YCbCr444ToRGBA(rgbaData.data(), rgbaStride, width, height,
-                              yPlane.data(), yPlaneStride,
-                              uPlane.data(), uvPlaneStride,
-                              vPlane.data(), uvPlaneStride, 0.299f, 0.114f, sparkyuv::YUV_RANGE_TV);
+                             yPlane.data(), yPlaneStride,
+                             uPlane.data(), uvPlaneStride,
+                             vPlane.data(), uvPlaneStride, 0.299f, 0.114f, sparkyuv::YUV_RANGE_TV);
 //    sparkyuv::YCbCr400ToRGBA(rgbaData.data(), rgbaStride, width, height,
 //                             yPlane.data(), yPlaneStride,0.299f, 0.114f, sparkyuv::YUV_RANGE_PC);
   });
@@ -192,8 +192,6 @@ int main() {
 //                       vPlane.data(), uvPlaneStride,  rgbaData.data(), rgbaStride, width, height);
     sparkyuv::FlipHorizontalRGBA(rgbaData.data(), rgbaStride, flipped2.data(), rgbaStride, width, height);
     rgbaData = flipped2;
-    sparkyuv::FlipHorizontalRGBA(rgbaData.data(), rgbaStride, flipped2.data(), rgbaStride, width, height);
-    rgbaData = flipped2;
 //    sparkyuv::YCbCr400ToRGBA(rgbaData.data(), rgbaStride, width, height,
 //                             yPlane.data(), yPlaneStride,0.299f, 0.114f, sparkyuv::YUV_RANGE_PC);
   });
@@ -326,10 +324,18 @@ int main() {
 //                              rgbaStride, inWidth,
 //                              inHeight);
 
+  int trnsStride = sizeof(uint8_t) * height * 4;
+  std::vector<uint8_t> transposed(trnsStride * width);
+  sparkyuv::TransposeCounterClockwiseRGBA(rgbaData.data(), rgbaStride, transposed.data(), trnsStride, width, height);
+
   aire::JPEGEncoder encoder(rgbaData.data(), rgbaStride, width, height);
   auto encoded = encoder.encode();
   saveVectorToFile(encoded, "jpeg.jpeg");
 
+  aire::JPEGEncoder encoderTrns(transposed.data(), trnsStride, height, width);
+  auto encodedTrns = encoderTrns.encode();
+  saveVectorToFile(encodedTrns, "jpeg_trns.jpeg");
+
   aire::JPEGEncoder encoderNV(rgbaNVData.data(), rgbaNVStride, width, height);
   auto encodedNV = encoderNV.encode();
   saveVectorToFile(encodedNV, "jpeg_nv.jpeg");