Develop Reference

Need to capture a still image during face detection with MLKit and Camera2

I'm developing a Face Detection feature with Camera2 and MLKit.
In the Developer Guide, in the Performance Tips part, they say to capture images in ImageFormat.YUV_420_888 format if using the Camera2 API, which is my case.
Then, in the Face Detector part, they recommend to use an image with dimensions of at least 480x360 pixels for faces recognition in real time, which is again my case.
Ok, let's go ! Here is my code, working well
private fun initializeCamera() = lifecycleScope.launch(Dispatchers.Main) {
// Open the selected camera
cameraDevice = openCamera(cameraManager, getCameraId(), cameraHandler)
val previewSize = if (isPortrait) {
Size(RECOMMANDED_CAPTURE_SIZE.width, RECOMMANDED_CAPTURE_SIZE.height)
} else {
Size(RECOMMANDED_CAPTURE_SIZE.height, RECOMMANDED_CAPTURE_SIZE.width)
}
// Initialize an image reader which will be used to display a preview
imageReader = ImageReader.newInstance(
previewSize.width, previewSize.height, ImageFormat.YUV_420_888, IMAGE_BUFFER_SIZE)
// Retrieve preview's frame and run detector
imageReader.setOnImageAvailableListener({ reader ->
lifecycleScope.launch(Dispatchers.Main) {
val image = reader.acquireNextImage()
logD { "Image available: ${image.timestamp}" }
faceDetector.runFaceDetection(image, getRotationCompensation())
image.close()
}
}, imageReaderHandler)
// Creates list of Surfaces where the camera will output frames
val targets = listOf(viewfinder.holder.surface, imageReader.surface)
// Start a capture session using our open camera and list of Surfaces where frames will go
session = createCaptureSession(cameraDevice, targets, cameraHandler)
val captureRequest = cameraDevice.createCaptureRequest(
CameraDevice.TEMPLATE_PREVIEW).apply {
addTarget(viewfinder.holder.surface)
addTarget(imageReader.surface)
}
// This will keep sending the capture request as frequently as possible until the
// session is torn down or session.stopRepeating() is called
session.setRepeatingRequest(captureRequest.build(), null, cameraHandler)
}
Now, I want to capture a still image...and this is my problem because, ideally, I want:
a full resolution image or, as least, bigger than 480x360
in JPEG format to be able to save it
The Camera2Basic sample demonstrates how to capture an image (samples for Video and SlowMotion are crashing) and MLKit sample uses the so old Camera API !! Fortunately, I've succeeded is mixing these samples to develop my feature but I'm failed to capture a still image with a different resolution.
I think I have to stop the preview session to recreate one for image capture but I'm not sure...
What I have done is the following but it's capturing images in 480x360:
session.stopRepeating()
// Unset the image reader listener
imageReader.setOnImageAvailableListener(null, null)
// Initialize an new image reader which will be used to capture still photos
// imageReader = ImageReader.newInstance(768, 1024, ImageFormat.JPEG, IMAGE_BUFFER_SIZE)
// Start a new image queue
val imageQueue = ArrayBlockingQueue<Image>(IMAGE_BUFFER_SIZE)
imageReader.setOnImageAvailableListener({ reader - >
val image = reader.acquireNextImage()
logD {"[Still] Image available in queue: ${image.timestamp}"}
if (imageQueue.size >= IMAGE_BUFFER_SIZE - 1) {
imageQueue.take().close()
}
imageQueue.add(image)
}, imageReaderHandler)
// Creates list of Surfaces where the camera will output frames
val targets = listOf(viewfinder.holder.surface, imageReader.surface)
val captureRequest = createStillCaptureRequest(cameraDevice, targets)
session.capture(captureRequest, object: CameraCaptureSession.CaptureCallback() {
override fun onCaptureCompleted(
session: CameraCaptureSession,
request: CaptureRequest,
result: TotalCaptureResult) {
super.onCaptureCompleted(session, request, result)
val resultTimestamp = result.get(CaptureResult.SENSOR_TIMESTAMP)
logD {"Capture result received: $resultTimestamp"}
// Set a timeout in case image captured is dropped from the pipeline
val exc = TimeoutException("Image dequeuing took too long")
val timeoutRunnable = Runnable {
continuation.resumeWithException(exc)
}
imageReaderHandler.postDelayed(timeoutRunnable, IMAGE_CAPTURE_TIMEOUT_MILLIS)
// Loop in the coroutine's context until an image with matching timestamp comes
// We need to launch the coroutine context again because the callback is done in
// the handler provided to the `capture` method, not in our coroutine context
# Suppress("BlockingMethodInNonBlockingContext")
lifecycleScope.launch(continuation.context) {
while (true) {
// Dequeue images while timestamps don't match
val image = imageQueue.take()
if (image.timestamp != resultTimestamp)
continue
logD {"Matching image dequeued: ${image.timestamp}"}
// Unset the image reader listener
imageReaderHandler.removeCallbacks(timeoutRunnable)
imageReader.setOnImageAvailableListener(null, null)
// Clear the queue of images, if there are left
while (imageQueue.size > 0) {
imageQueue.take()
.close()
}
// Compute EXIF orientation metadata
val rotation = getRotationCompensation()
val mirrored = cameraFacing == CameraCharacteristics.LENS_FACING_FRONT
val exifOrientation = computeExifOrientation(rotation, mirrored)
logE {"captured image size (w/h): ${image.width} / ${image.height}"}
// Build the result and resume progress
continuation.resume(CombinedCaptureResult(
image, result, exifOrientation, imageReader.imageFormat))
// There is no need to break out of the loop, this coroutine will suspend
}
}
}
}, cameraHandler)
}
If I uncomment the new ImageReader instanciation, I have this exception:
java.lang.IllegalArgumentException: CaptureRequest contains
unconfigured Input/Output Surface!
Can anyone help me ?

This IllegalArgumentException:
java.lang.IllegalArgumentException: CaptureRequest contains unconfigured Input/Output Surface!
... obviously refers to imageReader.surface.
Meanhile (with CameraX) this works different, see CameraFragment.kt ...
Issue #197: Firebase Face Detection Api issue while using cameraX API;
there might soon be a sample application matching your use case.

ImageReader is sensitive to the choice of format and/or combination of usage flags. The documentation points certain combinations of format may be unsupported. With some Android devices (perhaps some older phone models) you might find the IllegalArgumentException is not thrown using the JPEG format. But it doesn't help much - you want something versatile.
What I have done in the past is to use ImageFormat.YUV_420_888 format (this will be backed by the hardware and ImageReader implementation). This format does not contain pre-optimizations that prevent the application accessing the image via the internal array of planes. I notice you have used it successfully already in your initializeCamera() method.
You may then extract the image data from the frame you want
Image.Plane[] planes = img.getPlanes();
byte[] data = planes[0].getBuffer().array();
and then via a Bitmap create the still image using JPEG compression, PNG, or whichever encoding you choose.
ByteArrayOutputStream out = new ByteArrayOutputStream();
YuvImage yuvImage = new YuvImage(data, ImageFormat.NV21, width, height, null);
yuvImage.compressToJpeg(new Rect(0, 0, width, height), 100, out);
byte[] imageBytes = out.toByteArray();
Bitmap bitmap= BitmapFactory.decodeByteArray(imageBytes, 0, imageBytes.length);
ByteArrayOutputStream out2 = new ByteArrayOutputStream();
bitmap.compress(Bitmap.CompressFormat.JPEG, 75, out2);

Is there a way to delay a video feed from a cell phone camera in Unity?

I am making a project where you are supposed to be able to change the delay with which the feed from the cell phone camera is shown, so that people can see how their brains handle the delay/latency. I have managed to show the camera feed on a canvas that follows the camera around and fills the whole view of the Google Cardboard, but I am wondering how I could delay this video feed. Perhaps by using an image array of some sort?
I have tried searching for sollutions online, but I have come up short of an answer. I have tried a texture2D array, but the performance was really bad (I tried a modified version of this).
private bool camAvailable;
private WebCamTexture backCam;
private Texture defaultBackground;
public RawImage background;
public AspectRatioFitter fit;
// Start is called before the first frame update
void Start()
{
defaultBackground = background.texture;
WebCamDevice[] devices = WebCamTexture.devices;
if (devices.Length == 0 )
{
Debug.Log("No camera detected");
camAvailable = false;
return;
}
for (int i = 0; i < devices.Length; i++)
{
if (!devices[i].isFrontFacing)
{
backCam = new WebCamTexture(devices[i].name, Screen.width, Screen.height); // Used to find the correct camera
}
}
if (backCam == null)
{
Debug.Log("Unable to find back camera");
return;
}
backCam.Play();
background.texture = backCam;
camAvailable = true;
} //Tell me if this is not enough code, I don't really have a lot of experience in Unity, so I am unsure of how much is required for a minimal reproducible example
Should I use some sort of frame buffer or image/texture array for delaying the video? (Start "recording", wait a specified amount of time, start playing the "video" on the screen)
Thanks in advance!

How to record screen and take screenshots, using Android API?

Background
Android got a new API on Kitkat and Lollipop, to video capture the screen. You can do it either via the ADB tool or via code (starting from Lollipop).
Ever since the new API was out, many apps came to that use this feature, allowing to record the screen, and Microsoft even made its own Google-Now-On-tap competitor app.
Using ADB, you can use:
adb shell screenrecord /sdcard/video.mp4
You can even do it from within Android Studio itself.
The problem
I can't find any tutorial or explanation about how to do it using the API, meaning in code.
What I've found
The only place I've found is the documentations (here, under "Screen capturing and sharing"), telling me this:
Android 5.0 lets you add screen capturing and screen sharing
capabilities to your app with the new android.media.projection APIs.
This functionality is useful, for example, if you want to enable
screen sharing in a video conferencing app.
The new createVirtualDisplay() method allows your app to capture the
contents of the main screen (the default display) into a Surface
object, which your app can then send across the network. The API only
allows capturing non-secure screen content, and not system audio. To
begin screen capturing, your app must first request the user’s
permission by launching a screen capture dialog using an Intent
obtained through the createScreenCaptureIntent() method.
For an example of how to use the new APIs, see the MediaProjectionDemo
class in the sample project.
Thing is, I can't find any "MediaProjectionDemo" sample. Instead, I've found "Screen Capture" sample, but I don't understand how it works, as when I've run it, all I've seen is a blinking screen and I don't think it saves the video to a file. The sample seems very buggy.
The questions
How do I perform those actions using the new API:
start recording, optionally including audio (mic/speaker/both).
stop recording
take a screenshot instead of video.
Also, how do I customize it (resolution, requested fps, colors, time...)?

First step and the one which Ken White rightly suggested & which you may have already covered is the Example Code provided officially.
I have used their API earlier. I agree screenshot is pretty straight forward. But, screen recording is also under similar lines.
I will answer your questions in 3 sections and will wrap it up with a link. :)
1. Start Video Recording
private void startScreenRecord(final Intent intent) {
if (DEBUG) Log.v(TAG, "startScreenRecord:sMuxer=" + sMuxer);
synchronized(sSync) {
if (sMuxer == null) {
final int resultCode = intent.getIntExtra(EXTRA_RESULT_CODE, 0);
// get MediaProjection
final MediaProjection projection = mMediaProjectionManager.getMediaProjection(resultCode, intent);
if (projection != null) {
final DisplayMetrics metrics = getResources().getDisplayMetrics();
final int density = metrics.densityDpi;
if (DEBUG) Log.v(TAG, "startRecording:");
try {
sMuxer = new MediaMuxerWrapper(".mp4"); // if you record audio only, ".m4a" is also OK.
if (true) {
// for screen capturing
new MediaScreenEncoder(sMuxer, mMediaEncoderListener,
projection, metrics.widthPixels, metrics.heightPixels, density);
}
if (true) {
// for audio capturing
new MediaAudioEncoder(sMuxer, mMediaEncoderListener);
}
sMuxer.prepare();
sMuxer.startRecording();
} catch (final IOException e) {
Log.e(TAG, "startScreenRecord:", e);
}
}
}
}
}
2. Stop Video Recording
private void stopScreenRecord() {
if (DEBUG) Log.v(TAG, "stopScreenRecord:sMuxer=" + sMuxer);
synchronized(sSync) {
if (sMuxer != null) {
sMuxer.stopRecording();
sMuxer = null;
// you should not wait here
}
}
}
2.5. Pause and Resume Video Recording
private void pauseScreenRecord() {
synchronized(sSync) {
if (sMuxer != null) {
sMuxer.pauseRecording();
}
}
}
private void resumeScreenRecord() {
synchronized(sSync) {
if (sMuxer != null) {
sMuxer.resumeRecording();
}
}
}
Hope the code helps. Here is the original link to the code that I referred to and from which this implementation(Video recording) is also derived from.
3. Take screenshot Instead of Video
I think by default its easy to capture the image in bitmap format. You can still go ahead with MediaProjectionDemo example to capture screenshot.
[EDIT] : Code encrypt for screenshot
a. To create virtual display depending on device width / height
mImageReader = ImageReader.newInstance(mWidth, mHeight, PixelFormat.RGBA_8888, 2);
mVirtualDisplay = sMediaProjection.createVirtualDisplay(SCREENCAP_NAME, mWidth, mHeight, mDensity, VIRTUAL_DISPLAY_FLAGS, mImageReader.getSurface(), null, mHandler);
mImageReader.setOnImageAvailableListener(new ImageAvailableListener(), mHandler);
b. Then start the Screen Capture based on an intent or action-
startActivityForResult(mProjectionManager.createScreenCaptureIntent(), REQUEST_CODE);
Stop Media projection-
sMediaProjection.stop();
c. Then convert to image-
//Process the media capture
image = mImageReader.acquireLatestImage();
Image.Plane[] planes = image.getPlanes();
ByteBuffer buffer = planes[0].getBuffer();
int pixelStride = planes[0].getPixelStride();
int rowStride = planes[0].getRowStride();
int rowPadding = rowStride - pixelStride * mWidth;
//Create bitmap
bitmap = Bitmap.createBitmap(mWidth + rowPadding / pixelStride, mHeight, Bitmap.Config.ARGB_8888);
bitmap.copyPixelsFromBuffer(buffer);
//Write Bitmap to file in some path on the phone
fos = new FileOutputStream(STORE_DIRECTORY + "/myscreen_" + IMAGES_PRODUCED + ".png");
bitmap.compress(CompressFormat.PNG, 100, fos);
fos.close();
There are several implementations (full code) of Media Projection API available.
Some other links that can help you in your development-
Video Recording with MediaProjectionManager - website
android-ScreenCapture - github as per android developer's observations :)
screenrecorder - github
Capture and Record Android Screen using MediaProjection APIs - website
Hope it helps :) Happy coding and screen recording!
PS: Can you please tell me the Microsoft app you are talking about? I have not used it. Would like to try it :)

Grafika TextureMovieEncoder

I have recently been modifying Grafika's TextureMovieEncoder to create a recording of what I displayed onscreen: two Sprite2ds which are overlapping. Using the CameraCaptureActivity example as a reference point, I effectively ported what I created for my rendering thread into the TextureMovieEncoder but the output is jagged lines across the screen. I think I understand what's wrong, but I don't know how to fix it:
Some code:
private void prepareEncoder(EGLContext sharedContext, int width, int height, int bitRate,
File outputFile) {
try {
mVideoEncoder = new VideoEncoderCore(width, height, bitRate, outputFile);
} catch (IOException ioe) {
throw new RuntimeException(ioe);
}
mEglCore = new EglCore(sharedContext, EglCore.FLAG_RECORDABLE);
mInputWindowSurface = new WindowSurface(mEglCore, mVideoEncoder.getInputSurface(), true);
mInputWindowSurface.makeCurrent();
textureProgram = new Texture2dProgram(Texture2dProgram.ProgramType.TEXTURE_EXT);
backgroundDrawable = new Drawable2d(Drawable2d.Prefab.RECTANGLE);
backgroundRect = new Sprite2d(backgroundDrawable);
frontDrawable = new Drawable2d(Drawable2d.Prefab.RECTANGLE);
frontRect = new Sprite2d(frontDrawable);
backgroundRect.setTexture(backTextureId);
frontRect.setTexture(frontTextureId);
updateGeometry();
}
private void handleFrameAvailable(Transform transform, long timestampNanos) {
if (VERBOSE) Log.d(TAG, "handleFrameAvailable tr=" + transform);
mVideoEncoder.drainEncoder(false);
backgroundRect.draw(textureProgram, transform.movieMatrix);
frontRect.draw(textureProgram, transform.cameraMatrix);
mInputWindowSurface.setPresentationTime(timestampNanos);
mInputWindowSurface.swapBuffers();
}
I think the problem comes down to my lack of understanding of how to establish the right projection onto the WindowSurface for the VideoEncoder. In the Grafika example, FullFrameRect is used, which is easier since you can just use the identity matrix to stretch a given texture to the surface area. However, since I want to create the overlapping effect, I needed to use Sprite2d. Is the problem the shared EGLContext? Do I need to create a new one so that I can set the viewport to match the WindowSurface size? A bit lost on where to go from here.

Turns out the functionality of the code above was fine. The problem was the interaction between the TextureEncoder and the calling parent.
I was initializing the member variables backTextureId and frontTextureId after prepareEncoder and it was therefore recording garbage data into the output.

Android: decode JPGs in loop: random freezes

Here's another question for you :)
Basically i made a realtime streaming service, sending multiple jpegs to my android app, that decodes them as soon as he receives them.
// dIn is DataInputStream
// videoFeed is an ImageView
// bitmap is Bitmap
// hand is an Handler of the main thread
//CODE EXECUTED IN ANOTHER THERAD
byte[] inBuff = new byte[8];
byte[] imgBuff;
String inMsg;
while(socket.isConnected()) {
dIn.readFully(inBuff);
inMsg = new String(inBuff, "ASCII").trim();
int size = Integer.parseInt(inMsg);
imgBuff = new byte[size];
dIn.readFully(imgBuff);
out.write("SEND-NEXT-JPEG".getBytes("ASCII"));
bitmap = BitmapFactory.decodeByteArray(imgBuff, 0, size);
hand.post(setImage);
}
}
private Runnable setImage = new Runnable() {
#Override
public void run() {
videoFeed.setImageBitmap(bitmap);
}
};
The problem is that after about 10 or 20 jpegs are perfectly decoded in realtime, the app freezes for 400ms or so and then it continues to decode other 10/20 jpegs before another freeze...
I know that sending multiple jpegs it's not a good way for streaming video but i can only change the client (android app), not the server.
Do you have any idea for get a fluid video and avoid freezes? thanks!

Right now, you are using the three-parameter version of decodeByteArray(). Instead, switch to the four-parameter version, passing in a BitmapFactory.Options as the last value. On there, set inBitmap to be a Bitmap object that can be reused.
This requires you to maintain a small Bitmap object pool. It could be as simple as two Bitmap instances: the one that is presently being displayed and the one that you are preparing for the next "frame" of the video.
The catch is that, for API Level 18 and below, the Bitmap needs to be the same resolution (height and width in pixels). In your case, that's probably not a problem, as I would imagine that each of your bitmaps have the same resolution.