本教程是通过PaddlePaddle的PaddleSeg实现的,该开源库的地址为:http://github.com/PaddlPaddle/PaddleSeg ,使用开源库提供的预训练模型实现人物的图像语义分割,最终部署到Android应用上。关于如何在Android应用上使用PaddlePaddle模型,可以参考笔者的这篇文章《基于Paddle Lite在Android手机上实现图像分类》。
本教程开源代码地址:https://github.com/yeyupiaoling/ChangeHumanBackground
图像语义分割工具
首先编写一个可以在Android应用使用PaddlePaddle的图像语义分割模型的工具类,通过是这个PaddleLiteSegmentation
这个java工具类实现模型的加载和图像的预测。
首先是加载模型,获得一个预测器,其中inputShape
为图像的输入大小,NUM_THREADS
为使用线程数来预测图像,最高可以支持4个线程预测。
private PaddlePredictor paddlePredictor;
private Tensor inputTensor;
public static long[] inputShape = new long[]{1, 3, 513, 513};
private static final int NUM_THREADS = 4;
/**
* @param modelPath model path
*/
public PaddleLiteSegmentation(String modelPath) throws Exception {
File file = new File(modelPath);
if (!file.exists()) {
throw new Exception("model file is not exists!");
}
try {
MobileConfig config = new MobileConfig();
config.setModelFromFile(modelPath);
config.setThreads(NUM_THREADS);
config.setPowerMode(PowerMode.LITE_POWER_HIGH);
paddlePredictor = PaddlePredictor.createPaddlePredictor(config);
inputTensor = paddlePredictor.getInput(0);
inputTensor.resize(inputShape);
} catch (Exception e) {
e.printStackTrace();
throw new Exception("load model fail!");
}
}
在预测开始之前,写两个重构方法,这个我们这个工具不管是图片路径还是图像的Bitmap都可以实现语义分割了。
代码语言:javascript复制 public long[] predictImage(String image_path) throws Exception {
if (!new File(image_path).exists()) {
throw new Exception("image file is not exists!");
}
FileInputStream fis = new FileInputStream(image_path);
Bitmap bitmap = BitmapFactory.decodeStream(fis);
long[] result = predictImage(bitmap);
if (bitmap.isRecycled()) {
bitmap.recycle();
}
return result;
}
public long[] predictImage(Bitmap bitmap) throws Exception {
return predict(bitmap);
}
现在还不能预测,还需要对图像进行预处理的方法,预测器输入的是一个浮点数组,而不是一个Bitmap对象,所以需要这样的一个工具方法,把图像Bitmap转换为浮点数组,同时对图像进行预处理,如通道顺序的变换,有的模型还需要数据的标准化,但这里没有使用到。
代码语言:javascript复制 private float[] getScaledMatrix(Bitmap bitmap) {
int channels = (int) inputShape[1];
int width = (int) inputShape[2];
int height = (int) inputShape[3];
float[] inputData = new float[channels * width * height];
Bitmap rgbaImage = bitmap.copy(Bitmap.Config.ARGB_8888, true);
Bitmap scaleImage = Bitmap.createScaledBitmap(rgbaImage, width, height, true);
Log.d(TAG, scaleImage.getWidth() ", " scaleImage.getHeight());
if (channels == 3) {
// RGB = {0, 1, 2}, BGR = {2, 1, 0}
int[] channelIdx = new int[]{0, 1, 2};
int[] channelStride = new int[]{width * height, width * height * 2};
for (int y = 0; y < height; y ) {
for (int x = 0; x < width; x ) {
int color = scaleImage.getPixel(x, y);
float[] rgb = new float[]{(float) red(color), (float) green(color), (float) blue(color)};
inputData[y * width x] = rgb[channelIdx[0]];
inputData[y * width x channelStride[0]] = rgb[channelIdx[1]];
inputData[y * width x channelStride[1]] = rgb[channelIdx[2]];
}
}
} else if (channels == 1) {
for (int y = 0; y < height; y ) {
for (int x = 0; x < width; x ) {
int color = scaleImage.getPixel(x, y);
float gray = (float) (red(color) green(color) blue(color));
inputData[y * width x] = gray;
}
}
} else {
Log.e(TAG, "图片的通道数必须是1或者3");
}
return inputData;
}
最后就可以执行预测了,预测的结果是一个数组,它代表了整个图像的语义分割的情况,0的为背景,1的为人物。
代码语言:javascript复制 private long[] predict(Bitmap bmp) throws Exception {
float[] inputData = getScaledMatrix(bmp);
inputTensor.setData(inputData);
try {
paddlePredictor.run();
} catch (Exception e) {
throw new Exception("predict image fail! log:" e);
}
Tensor outputTensor = paddlePredictor.getOutput(0);
long[] output = outputTensor.getLongData();
long[] outputShape = outputTensor.shape();
Log.d(TAG, "结果shape:" Arrays.toString(outputShape));
return output;
}
实现人物背景更换
在MainActivity
中,程序加载的时候就从assets中把模型复制到缓存目录中,然后加载图像语义分割模型。
String segmentationModelPath = getCacheDir().getAbsolutePath() File.separator "model.nb";
Utils.copyFileFromAsset(MainActivity.this, "model.nb", segmentationModelPath);
try {
paddleLiteSegmentation = new PaddleLiteSegmentation(segmentationModelPath);
Toast.makeText(MainActivity.this, "模型加载成功!", Toast.LENGTH_SHORT).show();
Log.d(TAG, "模型加载成功!");
} catch (Exception e) {
Toast.makeText(MainActivity.this, "模型加载失败!", Toast.LENGTH_SHORT).show();
Log.d(TAG, "模型加载失败!");
e.printStackTrace();
finish();
}
创建几个按钮,来控制图片背景的更换。
代码语言:javascript复制// 获取控件
Button selectPicture = findViewById(R.id.select_picture);
Button selectBackground = findViewById(R.id.select_background);
Button savePicture = findViewById(R.id.save_picture);
imageView = findViewById(R.id.imageView);
selectPicture.setOnClickListener(new View.OnClickListener() {
@Override
public void onClick(View v) {
// 打开相册
Intent intent = new Intent(Intent.ACTION_PICK);
intent.setType("image/*");
startActivityForResult(intent, 0);
}
});
selectBackground.setOnClickListener(new View.OnClickListener() {
@Override
public void onClick(View v) {
if (resultPicture != null){
// 打开相册
Intent intent = new Intent(Intent.ACTION_PICK);
intent.setType("image/*");
startActivityForResult(intent, 1);
}else {
Toast.makeText(MainActivity.this, "先选择人物图片!", Toast.LENGTH_SHORT).show();
}
}
});
savePicture.setOnClickListener(new View.OnClickListener() {
@Override
public void onClick(View v) {
// 保持图片
String savePth = Utils.saveBitmap(mergeBitmap1);
if (savePth != null) {
Toast.makeText(MainActivity.this, "图片保存:" savePth, Toast.LENGTH_SHORT).show();
Log.d(TAG, "图片保存:" savePth);
} else {
Toast.makeText(MainActivity.this, "图片保存失败", Toast.LENGTH_SHORT).show();
Log.d(TAG, "图片保存失败");
}
}
});
首先需要选择包含人物的图片,这时就需要对图像进行预测,获取语义分割结果,然后将图像放大的跟原图像一样大小,并做这个临时的画布。
代码语言:javascript复制Uri image_uri = data.getData();
image_path = Utils.getPathFromURI(MainActivity.this, image_uri);
try {
// 预测图像
FileInputStream fis = new FileInputStream(image_path);
Bitmap b = BitmapFactory.decodeStream(fis);
long start = System.currentTimeMillis();
long[] result = paddleLiteSegmentation.predictImage(image_path);
long end = System.currentTimeMillis();
// 创建一个任务为全黑色,背景完全透明的图片
humanPicture = b.copy(Bitmap.Config.ARGB_8888, true);
final int[] colors_map = {0x00000000, 0xFF000000};
int[] objectColor = new int[result.length];
for (int i = 0; i < result.length; i ) {
objectColor[i] = colors_map[(int) result[i]];
}
Bitmap.Config config = humanPicture.getConfig();
Bitmap outputImage = Bitmap.createBitmap(objectColor, (int) PaddleLiteSegmentation.inputShape[2], (int) PaddleLiteSegmentation.inputShape[3], config);
resultPicture = Bitmap.createScaledBitmap(outputImage, humanPicture.getWidth(), humanPicture.getHeight(), true);
imageView.setImageBitmap(b);
Log.d(TAG, "预测时间:" (end - start) "ms");
} catch (Exception e) {
e.printStackTrace();
}
最后在这里实现人物背景的更换,
代码语言:javascript复制Uri image_uri = data.getData();
image_path = Utils.getPathFromURI(MainActivity.this, image_uri);
try {
FileInputStream fis = new FileInputStream(image_path);
changeBackgroundPicture = BitmapFactory.decodeStream(fis);
mergeBitmap1 = draw();
imageView.setImageBitmap(mergeBitmap1);
} catch (Exception e) {
e.printStackTrace();
}
// 实现换背景
public Bitmap draw() {
// 创建一个对应人物位置透明其他正常的背景图
Bitmap bgBitmap = Bitmap.createScaledBitmap(changeBackgroundPicture, resultPicture.getWidth(), resultPicture.getHeight(), true);
for (int y = 0; y < resultPicture.getHeight(); y ) {
for (int x = 0; x < resultPicture.getWidth(); x ) {
int color = resultPicture.getPixel(x, y);
int a = Color.alpha(color);
if (a == 255) {
bgBitmap.setPixel(x, y, Color.TRANSPARENT);
}
}
}
// 添加画布,保证透明
Bitmap bgBitmap2 = Bitmap.createBitmap(bgBitmap.getWidth(), bgBitmap.getHeight(), Bitmap.Config.ARGB_8888);
Canvas canvas1 = new Canvas(bgBitmap2);
canvas1.drawBitmap(bgBitmap, 0, 0, null);
return mergeBitmap(humanPicture, bgBitmap2);
}
// 合并两张图片
public static Bitmap mergeBitmap(Bitmap backBitmap, Bitmap frontBitmap) {
Bitmap bitmap = backBitmap.copy(Bitmap.Config.ARGB_8888, true);
Canvas canvas = new Canvas(bitmap);
Rect baseRect = new Rect(0, 0, backBitmap.getWidth(), backBitmap.getHeight());
Rect frontRect = new Rect(0, 0, frontBitmap.getWidth(), frontBitmap.getHeight());
canvas.drawBitmap(frontBitmap, frontRect, baseRect, null);
return bitmap;
}
实现的效果如下: