Develop Reference

Android MediaCodec realtime h264 encoding/decoding latency

I'm working with Android MediaCodec and use it for a realtime H264 encoding and decoding frames from camera. I use MediaCodec in synchronous manner and render the output to the Surface of decoder and everething works fine except that I have a long latency from a realtime, it takes 1.5-2 seconds and I'm very confused why is it so.
I measured a total time of encoding and decoding processes and it keeps around 50-65 milliseconds so I think the problem isn't in them.
I tried to change the configuration of the encoder but it didn't help and currently it configured like this:
val formatEncoder = MediaFormat.createVideoFormat("video/avc", 1920, 1080)
formatEncoder.setInteger(MediaFormat.KEY_FRAME_RATE, 30)
formatEncoder.setInteger(MediaFormat.KEY_I_FRAME_INTERVAL, 5)
formatEncoder.setInteger(MediaFormat.KEY_BIT_RATE, 1920 * 1080)
formatEncoder.setInteger(MediaFormat.KEY_COLOR_FORMAT, MediaCodecInfo.CodecCapabilities.COLOR_FormatSurface)
val encoder = MediaCodec.createEncoderByType("video/avc")
encoder.configure(formatEncoder, null, null, MediaCodec.CONFIGURE_FLAG_ENCODE)
val inputSurface = encoder.createInputSurface() // I use it to send frames from camera to encoder
encoder.start()
Changing the configuration of the decoder also didn't help me at all and currently I configured it like this:
val formatDecoder = MediaFormat.createVideoFormat("video/avc", 1920, 1080)
val decoder = MediaCodec.createDecoderByType("video/avc")
decoder.configure(formatDecoder , outputSurface, null, 0) // I use outputSurface to render decoded frames into it
decoder.start()
I use the following timeouts for waiting for available encoder/decoder buffers I tried to reduce their values but it didn't help me and I left them like this:
var TIMEOUT_IN_BUFFER = 10000L // microseconds
var TIMEOUT_OUT_BUFFER = 10000L // microseconds
Also I measured the time of consuming the inputSurface a frame and this time takes 0.03-0.05 milliseconds so it isn't a bottleneck. Actually I measured all the places where a bottleneck could be, but I wasn't found anything and I think the problem is in the encoder or decoder itself or in their configurations, or maybe I should use some special routine for sending frames to encoding/decoding..
I also tried to use HW accelerated codec and it's the only thing that helped me, when I use it the latency reduces to ~ 500-800 milliseconds but it still doesn't fit me for a realtime streaming.
It seems to me that the encoder or decoder buffers several frames before start displaying them on the surface and eventually it leads to the latency and if it really so then how can I disable bufferization or reduce the time of it?
Please help me I'm stucking on this problem for about half a year and have no idea how to reduce the latency, I'm sure that it's possible because popular apps like Telegram, Viber, WhatsApp etc. work fine and without latency so what's the secret here?
UPD 07.07.2021:
I still haven't found a solution to get rid of the latency. I've tried to change h264 profiles, increase and decrease I-frame inteval, bitrate, framerate, but result the same, the only thing that hepls a little to reduce the latency - downgrade the resolution from 1920x1080 to e.g. 640x480, but this "solution" doesn't suit me because I want to encode/decode a realtime video with 1920x1080 resolution.
UPD 08.07.2021:
I found out that if I change the values of TIMEOUT_IN_BUFFER and TIMEOUT_OUT_BUFFER from 10_000L to 100_000L it decreases the latency a bit but increases the delay of showing the first frame quite a lot after start encoding/decoding process.

It's possible your encoder is producing B frames -- bilinear interpolation frames. They increase quality and latency, and are great for movies. But no good for low-latency applications.
Key frames = I (interframes)
Predicted frames = P (difference from previous frames)
Interpolated frames = B
A sequence of frames including B frames might look like this:
IBBBPBBBPBBBPBBBI
11111111
12345678901234567
The encoder must encode each P frame, and the decoder must decode it, before the preceding B frames make any sense. So in this example the frames get encoded out of order like this:
1 5 2 3 4 9 6 7 8 13 10 11 12 17 17 13 14 15
In this example the decoder can't handle frame 2 until the encoder has sent frame 5.
On the other hand, this sequence without B frames allows coding and decoding the frames in order.
IPPPPPPPPPPIPPPPPPPPP
Try using the Constrained Baseline Profile setting. It's designed for low latency and low power use. It suppresses B frames. I think this works.
mediaFormat.setInteger(
"profile",
CodecProfileLevel.AVCProfileConstrainedBaseline);

I believe android h264 decoder have latency (at-least in most cases i've tried). Probably that's why android developers added PARAMETER_KEY_LOW_LATENCY from API level 30.
However I could decrease the delay some frames by querying for the output some more times.
Reason: no idea. It's just result of boring trial and errors
int inputIndex = m_codec.dequeueInputBuffer(-1);// Pass in -1 here bc we don't have a playback time reference
if (inputIndex >= 0) {
ByteBuffer buffer;
if (android.os.Build.VERSION.SDK_INT >= android.os.Build.VERSION_CODES.LOLLIPOP) {
buffer = m_codec.getInputBuffer(inputIndex);
} else {
ByteBuffer[] bbuf = m_codec.getInputBuffers();
buffer = bbuf[inputIndex];
}
buffer.put(frame);
// tell the decoder to process the frame
m_codec.queueInputBuffer(inputIndex, 0, frame.length, 0, 0);
}
MediaCodec.BufferInfo info = new MediaCodec.BufferInfo();
int outputIndex = m_codec.dequeueOutputBuffer(info, 0);
if (outputIndex >= 0) {
m_codec.releaseOutputBuffer(outputIndex, true);
}
outputIndex = m_codec.dequeueOutputBuffer(info, 0);
if (outputIndex >= 0) {
m_codec.releaseOutputBuffer(outputIndex, true);
}
outputIndex = m_codec.dequeueOutputBuffer(info, 0);
if (outputIndex >= 0) {
m_codec.releaseOutputBuffer(outputIndex, true);
}

You need to configure customized(or KEY_LOW_LATENCY if it is supported) low latency parameters for different cpu venders. It is a common problem for android phone.
Check this code https://github.com/moonlight-stream/moonlight-android/blob/master/app/src/main/java/com/limelight/binding/video/MediaCodecHelper.java

android - MediaCodec video only decoder framerate control

Refer to this link, I just add a simple delay when output buffer available:
ByteBuffer buffer = outputBuffers[outIndex];
Log.v("DecodeActivity", "We can't use this buffer but render it due to the API limit, " + buffer);
// We use a very simple clock to keep the video FPS, or the video
// playback will be too fast
while (info.presentationTimeUs / 1000 > System.currentTimeMillis() - startMs) {
try {
sleep(10);
} catch (InterruptedException e) {
e.printStackTrace();
break;
}
}
decoder.releaseOutputBuffer(outIndex, true);
But when I feed a 25fps video only frames, the decoded video looks like only 10fps (many frames looks like dropped).
But if I add a frameconut to check the fps, it's really 25fps, and if add MediaMuxer to mux the frames in input buffer, it playbacks fine, which means frames actually not been dropped.
So it's wired why fames there but not show on screen, but if I remove the delay, the playback will be very quick (almost 50fps).

Just found issue caused by TextureView, after change TextureView to SurfaceView, it works fine now.
But still not clear why TextureView performance is so bad.

Android camera2 jpeg framerate

I am trying to save image sequences with fixed framerates (preferably up to 30) on an android device with FULL capability for camera2 (Galaxy S7), but I am unable to a) get a steady framerate, b) reach even 20fps (with jpeg encoding). I already included the suggestions from Android camera2 capture burst is too slow.
The minimum frame duration for JPEG is 33.33 milliseconds (for resolutions below 1920x1080) according to
characteristics.get(CameraCharacteristics.SCALER_STREAM_CONFIGURATION_MAP).getOutputMinFrameDuration(ImageFormat.JPEG, size);
and the stallduration is 0ms for every size (similar for YUV_420_888).
My capture builder looks as follows:
captureBuilder.set(CaptureRequest.CONTROL_AE_MODE, CONTROL_AE_MODE_OFF);
captureBuilder.set(CaptureRequest.SENSOR_EXPOSURE_TIME, _exp_time);
captureBuilder.set(CaptureRequest.CONTROL_AE_LOCK, true);
captureBuilder.set(CaptureRequest.SENSOR_SENSITIVITY, _iso_value);
captureBuilder.set(CaptureRequest.LENS_FOCUS_DISTANCE, _foc_dist);
captureBuilder.set(CaptureRequest.CONTROL_AF_MODE, CONTROL_AF_MODE_OFF);
captureBuilder.set(CaptureRequest.CONTROL_AWB_MODE, _wb_value);
// https://stackoverflow.com/questions/29265126/android-camera2-capture-burst-is-too-slow
captureBuilder.set(CaptureRequest.EDGE_MODE,CaptureRequest.EDGE_MODE_OFF);
captureBuilder.set(CaptureRequest.COLOR_CORRECTION_ABERRATION_MODE, CaptureRequest.COLOR_CORRECTION_ABERRATION_MODE_OFF);
captureBuilder.set(CaptureRequest.NOISE_REDUCTION_MODE, CaptureRequest.NOISE_REDUCTION_MODE_OFF);
captureBuilder.set(CaptureRequest.CONTROL_AF_TRIGGER, CaptureRequest.CONTROL_AF_TRIGGER_CANCEL);
// Orientation
int rotation = getWindowManager().getDefaultDisplay().getRotation();
captureBuilder.set(CaptureRequest.JPEG_ORIENTATION,ORIENTATIONS.get(rotation));
Focus distance is set to 0.0 (inf), iso is set to 100, exposure-time 5ms. Whitebalance can be set to OFF/AUTO/ANY VALUE, it does not impact the times below.
I start the capture session with the following command:
session.setRepeatingRequest(_capReq.build(), captureListener, mBackgroundHandler);
Note: It does not make a difference if I request RepeatingRequest or RepeatingBurst..
In the preview (only texture surface attached), everything is at 30fps.
However, as soon as I attach an image reader (listener running on HandlerThread) which I instantiate like follows (without saving, only measuring time between frames):
reader = ImageReader.newInstance(_img_width, _img_height, ImageFormat.JPEG, 2);
reader.setOnImageAvailableListener(readerListener, mBackgroundHandler);
With time-measuring code:
ImageReader.OnImageAvailableListener readerListener = new ImageReader.OnImageAvailableListener() {
#Override
public void onImageAvailable(ImageReader myreader) {
Image image = null;
image = myreader.acquireNextImage();
if (image == null) {
return;
}
long curr = image.getTimestamp();
Log.d("curr- _last_ts", "" + ((curr - last_ts) / 1000000) + " ms");
last_ts = curr;
image.close();
}
}
I get periodically repeating time differences like this:
99 ms - 66 ms - 66 ms - 99 ms - 66 ms - 66 ms ...
I do not understand why these take double or triple the time that the stream configuration map advertised for jpeg? The exposure time is well below the frame duration of 33ms. Is there some other internal processing happening that I am not aware of?
I tried the same for the YUV_420_888 format, which resulted in constant time-differences of 33ms. The problem I have here is that the cellphone lacks the bandwidth to store the images fast enough (I tried the method described in How to save a YUV_420_888 image?). If you know of any method to compress or encode these images fast enough myself, please let me know.
Edit: From the documentation of getOutputStallDuration: "In other words, using a repeating YUV request would result in a steady frame rate (let's say it's 30 FPS). If a single JPEG request is submitted periodically, the frame rate will stay at 30 FPS (as long as we wait for the previous JPEG to return each time). If we try to submit a repeating YUV + JPEG request, then the frame rate will drop from 30 FPS." Does this imply that I need to periodically request a single capture()?
Edit2: From https://developer.android.com/reference/android/hardware/camera2/CaptureRequest.html: "The necessary information for the application, given the model above, is provided via the android.scaler.streamConfigurationMap field using getOutputMinFrameDuration(int, Size). These are used to determine the maximum frame rate / minimum frame duration that is possible for a given stream configuration.
Specifically, the application can use the following rules to determine the minimum frame duration it can request from the camera device:
Let the set of currently configured input/output streams be called S.
Find the minimum frame durations for each stream in S, by looking it up in android.scaler.streamConfigurationMap using getOutputMinFrameDuration(int, Size) (with its respective size/format). Let this set of frame durations be called F.
For any given request R, the minimum frame duration allowed for R is the maximum out of all values in F. Let the streams used in R be called S_r.
If none of the streams in S_r have a stall time (listed in getOutputStallDuration(int, Size) using its respective size/format), then the frame duration in F determines the steady state frame rate that the application will get if it uses R as a repeating request."

The JPEG output is by way not the fastest way to fetch frames. You can accomplish this a lot faster by drawing the frames directly onto a Quad using OpenGL.
For burst capture, a faster solution would be capturing the images to RAM without encoding them, then encoding and saving them asynchronously.
On this website you can find a lot of excellent code related to android multimedia in general.
This specific program uses OpenGL to fetch the pixel data from an MPEG video. It's not difficult to use the camera as input instead of a video. You can basically use the texture used in the CodecOutputSurface class from the mentioned program as output texture for your capture request.

A possible solution I found consists of using and dumping YUV without encoding it as JPEG in combination with a micro Sd-card that is able to save up to 95Mb per second. (I had the misconception that YUV images would be larger, so with a cellphone that has full support for the camera2-pipeline, the write speed should be the limiting factor.
With this setup, I was able to achieve the following stable rates:
1920x1080, 15fps (approx. 4Mb * 15 == 60Mb/sec)
960x720, 30fps. (approx. 1.5Mb * 30 == 45Mb/sec)
I then encode the images offline from YUV to PNG using a python script.

Controlling Frame Rate of VirtualDisplay

I'm writing an Android application, and in it, I have a VirtualDisplay to mirror what is on the screen and I then send the frames from the screen to an instance of a MediaCodec. It works, but, I want to add a way of specifying the FPS of the encoded video, but I'm unsure how to do so.
From what I've read and experimented with, dropping encoded frames (based on the presentation times) doesn't work well as it ends up with blocky/artifact ridden video as opposed to a smooth video at a lower framerate. Other reading suggests that the only way to do what I want (limit the FPS) would be to limit the incoming FPS to the MediaCodec, but the VirtualDisplay just receives a Surface which is constructed from the MediaCodec as below
mSurface = <instance of MediaCodec>.createInputSurface();
mVirtualDisplay = mMediaProjection.createVirtualDisplay(
"MyDisplay",
screenWidth,
screenHeight,
screenDensity,
DisplayManager.VIRTUAL_DISPLAY_FLAG_AUTO_MIRROR,
mSurface,
null,
null);
I've also tried subclassing Surface and limit the frames that are fed to the MediaCodec via the unlockCanvasAndPost(Canvas canvas) but the function never seems to be called on my instance, so, there may be some weirdness in how I extended Surface and the interaction with the Parcel as writeToParcel function is called on my instance, but that is the only function that is called in my instance (that I can tell).
Other reading suggests that I can go from encoder -> decoder -> encoder and limit the rate in which the second encoder is fed frames, but that's a lot of extra computation that I'd rather not do if I can avoid it.
Has anyone successfully limited the rate at which a VirtualDisplay feeds its Surface? Any help would be greatly appreciated!

Starting off with what you can't do...
You can't drop content from the encoded stream. Most of the frames in the encoded stream are essentially "diffs" from other frames. Without knowing how the frames interact, you can't safely drop content, and will end up with that corrupted macroblock look.
You can't specify the frame rate to the MediaCodec encoder. It might stuff that into metadata somewhere, but the only thing that really matters to the codec is the frames you're feeding into it, and the presentation time stamps associated with each frame. The encoder will not drop frames.
You can't do anything useful by subclassing Surface. The Canvas operations are only used for software rendering, which is unrelated to feeding in frames from a camera or virtual display.
What you can do is send the frames to an intermediate Surface, and then choose whether or not to forward them to the MediaCodec's input Surface. One approach would be to create a SurfaceTexture, construct a Surface from it, and pass that to the virtual display. When the SurfaceTexture's frame-available callback fires, you either ignore it, or render the texture onto the MediaCodec input Surface with GLES.
Various examples can be found in Grafika and on bigflake, none of which are an exact fit, but all of the necessary EGL and GLES classes are there.

You can reference the code sample from saki4510t's ScreenRecordingSample or RyanRQ's ScreenRecoder, they are all use the additional EGL Texture between the virtual display and media encoder, and the first one can keep at least 15 fps for the output video. You can search the keyword createVirtualDisplay from their code base for more details.

How to handle the PTS correctly using Android AudioRecord and MediaCodec as audio encoder?

I'm using AudioRecord to record the audio stream during a camera capturing process on Android device.
Since I want to process the frame data and handle audio/video samples, I do not use MediaRecorder.
I run AudioRecord in another thread with the calling of read() to gather the raw audio data.
Once I get a data stream, I feed them into an MediaCodec configured as an AAC audio encoder.
Here are some of my codes about the audio recorder / encoder:
m_encode_audio_mime = "audio/mp4a-latm";
m_audio_sample_rate = 44100;
m_audio_channels = AudioFormat.CHANNEL_IN_MONO;
m_audio_channel_count = (m_audio_channels == AudioFormat.CHANNEL_IN_MONO ? 1 : 2);
int audio_bit_rate = 64000;
int audio_data_format = AudioFormat.ENCODING_PCM_16BIT;
m_audio_buffer_size = AudioRecord.getMinBufferSize(m_audio_sample_rate, m_audio_channels, audio_data_format) * 2;
m_audio_recorder = new AudioRecord(MediaRecorder.AudioSource.MIC, m_audio_sample_rate,
m_audio_channels, audio_data_format, m_audio_buffer_size);
m_audio_encoder = MediaCodec.createEncoderByType(m_encode_audio_mime);
MediaFormat audio_format = new MediaFormat();
audio_format.setString(MediaFormat.KEY_MIME, m_encode_audio_mime);
audio_format.setInteger(MediaFormat.KEY_BIT_RATE, audio_bit_rate);
audio_format.setInteger(MediaFormat.KEY_CHANNEL_COUNT, m_audio_channel_count);
audio_format.setInteger(MediaFormat.KEY_SAMPLE_RATE, m_audio_sample_rate);
audio_format.setInteger(MediaFormat.KEY_AAC_PROFILE, MediaCodecInfo.CodecProfileLevel.AACObjectLC);
audio_format.setInteger(MediaFormat.KEY_MAX_INPUT_SIZE, m_audio_buffer_size);
m_audio_encoder.configure(audio_format, null, null, MediaCodec.CONFIGURE_FLAG_ENCODE);
I found that the first time of AudioRecord.read() takes longer time to return, while the successive read() have time intervals that are more close to the real time of audio data.
For example, my audio format is 44100Hz 16Bit 1Channel, and the buffer size of AudioRecord is 16384, so a full buffer means 185.76 ms. When I record the system time for each call of read() and subtracting them from a base time, I get the following sequence:
time before each read(): 0ms, 345ms, 543ms, 692ms, 891ms, 1093ms, 1244ms, ...
I feed these raw data to the audio encoder with the above time values as PTS, and the encoder outputs encoded audio samples with the following PTS:
encoder output PTS: 0ms, 185ms, 371ms, 557ms, 743ms, 928ms, ...
It looks like that the encoder treats each part of data as having the same time period. I believe that the encoder works correctly since I give it raw data with the same size (16384) every time. However, if I use the encoder output PTS as the input of muxer, I'll get a video with audio content being faster then video content.
I want to ask that:
Is it expected that the first time of AudioRecord.read() blocks longer? I'm sure that the function call takes more than 300ms while it only records 16384 bytes as 186ms. Is this also an issue that depends on device / Android version?
What should I do to achieve audio/video synchronization? I have a workaround to measure the delay time of the first call of read(), then shift the PTS of audio samples by the delay. Is there another better way to handle this?

Convert the mono input to stereo. I was pulling my hair out for some time before I realised the AAC encoder exposed by MediaCoder only works with stereo input.