Implement timestamp translation/filter in VideoCapturer.

webrtc/media/base/videocapturer.cc

 /*
  *  Copyright (c) 2010 The WebRTC project authors. All Rights Reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS.  All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
@@ skipping 41 matching lines @@
   if (!size || data_size == CapturedFrame::kUnknownDataSize) {
     return false;
   }
   *size = data_size;
   return true;
 }
 
 /////////////////////////////////////////////////////////////////////
 // Implementation of class VideoCapturer
 /////////////////////////////////////////////////////////////////////
-VideoCapturer::VideoCapturer() : apply_rotation_(false) {
+VideoCapturer::VideoCapturer()
+    : apply_rotation_(false), frames_seen_(0), offset_us_(0) {
   thread_checker_.DetachFromThread();
   Construct();
 }
 
 void VideoCapturer::Construct() {
   enable_camera_list_ = false;
   capture_state_ = CS_STOPPED;
   SignalFrameCaptured.connect(this, &VideoCapturer::OnFrameCaptured);
   scaled_width_ = 0;
   scaled_height_ = 0;
@@ skipping 134 matching lines @@
   if (frame_factory_) {
     frame_factory_->SetApplyRotation(apply_rotation_);
   }
 
   if (video_adapter()) {
     video_adapter()->OnResolutionRequest(wants.max_pixel_count,
                                          wants.max_pixel_count_step_up);
   }
 }
 
+void VideoCapturer::UpdateOffset(int64_t capture_time_us) {
+  // Estimate the offset between system monotonic time and the capture
+  // time from the camera. The camera is assumed to provide more
+  // accurate timestamps than we can do here. But the camera may use
+  // its own free-running clock with a large offset and a small drift
+  // compared to the system clock. So the model is basically
+  //
+  //   y_k = c_0 + c_1 x_k + v_k
+  //
+  // where x_k is the camera timestamp, believed to be accurate in its
+  // own scale. y_k is our reading of the system clock. v_k is the
+  // measurement noise, i.e., the delay from frame capture until we
+  // get here and read the clock.
+  //
+  // It's possible to do (weighted) least-squares estimation of both
+  // c_0 and c_1. Then we get the constants as c_1 = Cov(x,y) /
+  // Var(x), and c_0 = mean(y) - c_1 mean(x). Substituting this c_0,
+  // we can rearrange the model as
+  //
+  //   y_k = mean(y) + (x_k - mean(x)) + (c_1 - 1) (x_k - mean(x)) + v_k
+  //
+  // Now if we use a weighted average which gradually forgets old
+  // values, x_k - mean(x) is bounded, of the same order as the time
+  // constant (and close to constant for a steady frame rate). In
+  // addition, the frequency error |c_1 - 1| should be small. Cameras
+  // with a frequency error up to 3000 ppm (3 ms drift per second)
+  // have been observed, but frequency errors below 100 ppm could be
+  // expected of any cheap crystal.
+  //
+  // Bottom line is that we ignore the c_1 term, and use only the estimator
+  //
+  //    x_k + mean(y-x)
+  //
+  // where mean is plain averaging for initial samples, followed by
+  // exponential averaging.
+
+  // TODO(nisse): Don't read the clock here, instead let the caller
+  // pass in the current system time? Useful for testing, or if the
+  // application reads the system clock earlier.

[pthatcher1, 2016/05/27 22:47:44]

You could pass in a cricket::ClockInterface and use a cricket::FakeClock.  But
passing in a current time into UpdateOffset sounds cleaner to me anyway.

+  int64_t diff_us = rtc::TimeMicros() - capture_time_us;
+
+  // We also try to detect if the camera timestamp actually is using
+  // the system monotonic clock (a common case, for cameras without
+  // builtin timestamping). In this case, we aim to keep the
+  // timestamps as if, i.e., set the offset to zero. In case the
+  // camera clock is drifting, and by chance the offset is crossing
+  // zero, this hack will only cause a small dent in the otherwise
+  // linear offset curve, temporarily forcing it closer to zero.
+  static const int64_t kDelayLimitUs = 50000;
+  if (diff_us > 0 && diff_us < kDelayLimitUs)
+    diff_us = 0;

[pthatcher1, 2016/05/27 22:47:44]

{}s please

[nisse-webrtc, 2016/05/30 08:57:29]

Done.

+
+  // TODO(nisse): Do we need to detect jumps in the camera clock?
+  // E.g., if the camera is somehow reset mid-stream? We could check
+  // if abs(diff_us - offset_us) > 500ms or so, and in this case reset
+  // frames_seen_ to zero.
+  static const unsigned kWindowSize = 100;
+  if (frames_seen_ < kWindowSize)
+    ++frames_seen_;

[pthatcher1, 2016/05/27 22:47:44]

{}s here too

[nisse-webrtc, 2016/05/30 08:57:29]

Done.

+
+  offset_us_ += (diff_us - offset_us_) / frames_seen_;

[pthatcher1, 2016/05/27 22:47:44]

Oh wow.  I'm going to have to come back to this when I have more time.

+}
+
 bool VideoCapturer::AdaptFrame(int width,
                                int height,
-                               // TODO(nisse): Switch to us unit.
+                               // TODO(nisse): Switch to us unit. In
+                               // progress in different cl.
                                int64_t capture_time_ns,
                                int* out_width,
                                int* out_height,
                                int* crop_width,
                                int* crop_height,
                                int* crop_x,
-                               int* crop_y) {
+                               int* crop_y,
+                               int64_t* time_us) {
+  int64_t capture_time_us = capture_time_ns / rtc::kNumNanosecsPerMicrosec;
+  UpdateOffset(capture_time_us);
+
   if (!broadcaster_.frame_wanted()) {
     return false;
   }
 
   if (enable_video_adapter_ && !IsScreencast()) {
     if (!video_adapter_.AdaptFrameResolution(
             width, height, capture_time_ns,
             crop_width, crop_height, out_width, out_height)) {
       // VideoAdapter dropped the frame.
       return false;
     }
     *crop_x = (width - *crop_width) / 2;
     *crop_y = (height - *crop_height) / 2;
   } else {
     *out_width = width;
     *out_height = height;
     *crop_width = width;
     *crop_height = height;
     *crop_x = 0;
     *crop_y = 0;
   }
+  *time_us = capture_time_us + offset_us_;
   return true;
 }
 
 void VideoCapturer::OnFrameCaptured(VideoCapturer*,
                                     const CapturedFrame* captured_frame) {
   int out_width;
   int out_height;
   int crop_width;
   int crop_height;
   int crop_x;
   int crop_y;
+  int64_t time_us;
 
   if (!AdaptFrame(captured_frame->width, captured_frame->height,
                   captured_frame->time_stamp,
                   &out_width, &out_height,
-                  &crop_width, &crop_height, &crop_x, &crop_y)) {
+                  &crop_width, &crop_height, &crop_x, &crop_y, &time_us)) {
     return;
   }
 
   if (!frame_factory_) {
     LOG(LS_ERROR) << "No video frame factory.";
     return;
   }
 
+  frame_factory_->SetTimestampOffset(offset_us_);
+
   // TODO(nisse): Reorganize frame factory methods. crop_x and crop_y
   // are ignored for now.
   std::unique_ptr<VideoFrame> adapted_frame(frame_factory_->CreateAliasedFrame(
       captured_frame, crop_width, crop_height, out_width, out_height));
 
   if (!adapted_frame) {
     // TODO(fbarchard): LOG more information about captured frame attributes.
     LOG(LS_ERROR) << "Couldn't convert to I420! "
                   << "From " << ToString(captured_frame) << " To "
                   << out_width << " x " << out_height;
@@ skipping 151 matching lines @@
 void VideoCapturer::UpdateInputSize(int width, int height) {
   // Update stats protected from fetches from different thread.
   rtc::CritScope cs(&frame_stats_crit_);
 
   input_size_valid_ = true;
   input_width_ = width;
   input_height_ = height;
 }
 
 }  // namespace cricket