update README.md

README.md

@@ -8,16 +8,44 @@ When you're attempting to get an `Image` from `ImageReader` or `ImageAnalysis.An
 
 **Solution**:
 
-This library carefully merges 3 buffers into one with respect to all strides. As a result, you receive YUV type (`NV21` or `YUV_420`) and `ByteBuffer` to pass into OpenCV or a neural network framework.
+This library carefully merges 3 buffers into one with respect to all strides. As a result, you receive YUV type (`NV21` or `I420`) and `ByteBuffer` to pass into OpenCV or a neural network framework.
 
 The whole library is a single file, you can just copy [Yuv.java](yuv2buf/src/main/java/ru/gordinmitya/yuv2buf/Yuv.java) into your project.
 
+**Usage**
+
+```kotlin
+private var reuseBuffer: ByteBuffer? = null
+
+fun convert(image: ImageProxy): Pair<Bitmap, Long> {
+
+ val converted = Yuv.toBuffer(image)
+ // OR pass existing DirectBuffer for reuse
+ val converted = Yuv.toBuffer(image, reuseBuffer)
+ reuseBuffer = converted.buffer
+
+ val format = when (converted.type) {
+ Yuv.Type.YUV_I420 -> Imgproc.COLOR_YUV2RGBA_I420
+ Yuv.Type.YUV_NV21 -> Imgproc.COLOR_YUV2RGBA_NV21
+ }
+
+ // process with one of converters
+}
+```
+
+**Converters** 
+
+1. [OpenCVConverter.kt](app/src/main/java/ru/gordinmitya/yuv2buf_demo/OpenCVConverter.kt) - the fastest. If your goal is to get Mat consider this method from [OpenCV](https://github.com/opencv/opencv/blob/master/modules/java/generator/android-21/java/org/opencv/android/JavaCamera2View.java#L344).
+2. [RenderScriptConverter.kt](app/src/main/java/ru/gordinmitya/yuv2buf_demo/RenderScriptConverter.kt) - built-in, no additional libraries required.
+3. [MNNConverter.kt](app/src/main/java/ru/gordinmitya/yuv2buf_demo/MNNConverter.kt) - if your goal is futher processing with neural network.
+
 **Alternatives**
 
 1. (For OpenCV users) Copy private method from OpenCV camera implementation: [JavaCamera2View, Mat rgba()](https://github.com/opencv/opencv/blob/master/modules/java/generator/android-21/java/org/opencv/android/JavaCamera2View.java#L344).
 2. Capture image from camera directly to [RenderScript Allocation.getSurface()](https://developer.android.com/reference/android/renderscript/Allocation#getSurface());
-3. Switch back to CameraApi 1 [(some trade offs)](https://github.com/tensorflow/tensorflow/issues/22620);
-4. Manipulate pixels manually as it has done in TFLite demo [ImageUtils](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/android/src/org/tensorflow/demo/env/ImageUtils.java#L161).
+3. RenderScript implementation in [android/camera-samples](https://github.com/android/camera-samples/blob/3730442b49189f76a1083a98f3acf3f5f09222a3/CameraUtils/lib/src/main/java/com/example/android/camera/utils/YuvToRgbConverter.kt) repo.
+4. Switch back to CameraApi 1 [(some trade offs)](https://github.com/tensorflow/tensorflow/issues/22620);
+5. Manipulate pixels manually as it has done in TFLite demo [ImageUtils](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/android/src/org/tensorflow/demo/env/ImageUtils.java#L161).
  However, even with [C++ implementation](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/android/jni/yuv2rgb.cc#L61)
  it's ridiculously slow. ~50ms for image about 1280x720 on Snapdragon 855;
 
@@ -28,6 +56,7 @@ The whole library is a single file, you can just copy [Yuv.java](yuv2buf/src/mai
 - [x] write unit tests;
 - [x] add RenderScript example;
 - [x] add OpenCV example;
+- [x] add MNN example;
+- [ ] publish to GooglePlay;
 - [ ] publish to jcenter;
-- [ ] add MNN example;
 - [ ] add TFLite example.

app/build.gradle

@@ -6,6 +6,8 @@ android {
  compileSdkVersion 29
  buildToolsVersion "29.0.3"
 
+ ndkVersion "21.3.6528147"
+
  defaultConfig {
  applicationId "ru.gordinmitya.yuv2buf_demo"
  minSdkVersion 21