Implement timestamp translation/filter in VideoCapturer.

webrtc/media/base/videocapturer.cc

Index: webrtc/media/base/videocapturer.cc
diff --git a/webrtc/media/base/videocapturer.cc b/webrtc/media/base/videocapturer.cc
index 96a605585509220b12875ecf993fc5010fe5e1cc..820d4200edea4dcf118688162f8975e7ef7f32d2 100644
--- a/webrtc/media/base/videocapturer.cc
+++ b/webrtc/media/base/videocapturer.cc
@@ -59,7 +59,8 @@ bool CapturedFrame::GetDataSize(uint32_t* size) const {
 /////////////////////////////////////////////////////////////////////
 // Implementation of class VideoCapturer
 /////////////////////////////////////////////////////////////////////
-VideoCapturer::VideoCapturer() : apply_rotation_(false) {
+VideoCapturer::VideoCapturer()
+    : apply_rotation_(false), frames_seen_(0), offset_us_(0) {
   thread_checker_.DetachFromThread();
   Construct();
 }
@@ -214,23 +215,113 @@ void VideoCapturer::OnSinkWantsChanged(const rtc::VideoSinkWants& wants) {
   }
 }
 
+void VideoCapturer::UpdateOffset(int64_t camera_time_us,
+                                 int64_t system_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 get from the system time. 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 the
+  // system clock was read.
+  //
+  // 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.
+
+  // The input for averaging, y_k - x_k in the above notation.
+  int64_t diff_us = system_time_us - camera_time_us;
+  // The deviation from the current average.
+  int64_t error_us = diff_us - offset_us_;
+
+  // If the current difference is far from the currently estimated
+  // offset, the filter is reset. This could happen, e.g., if the
+  // camera clock is reset, or cameras are plugged in and out, or if
+  // the application process is temporarily suspended. The limit of
+  // 300 ms should make this unlikely in normal operation, and at the
+  // same time, converging gradually rather than resetting the filter
+  // should be tolerable for jumps in camera time below this
+  // threshold.
+  static const int64_t kResetLimitUs = 300000;
+  if (std::abs(error_us) > kResetLimitUs) {
+    LOG(LS_INFO) << "Resetting timestamp translation after averaging "
+                 << frames_seen_ << " frames. Old offset: "
+                 << offset_us_ << ", new offset: " << diff_us;
+    frames_seen_ = 0;
+    prev_translated_time_us_ = rtc::Optional<int64_t>();
+  }
+
+  static const unsigned kWindowSize = 100;
+  if (frames_seen_ < kWindowSize) {
+    ++frames_seen_;
+  }
+  offset_us_ += error_us / frames_seen_;
+}
+
+int64_t VideoCapturer::ClipTimestamp(int64_t time_us, int64_t system_time_us) {
+  // Make timestamps monotonic.
+  if (!prev_translated_time_us_) {
+    // Initialize.
+    clip_bias_us_ = 0;

[nisse-webrtc, 2016/06/20 11:47:49]

It might make sense to initialize it to a positive value (say, 10 ms or so), to
make it constant in the common case.

+  } else if (time_us < *prev_translated_time_us_) {
+    time_us = *prev_translated_time_us_;
+  }
+
+  // Clip to make sure we don't produce time stamps in the future.
+  time_us -= clip_bias_us_;
+  if (time_us > system_time_us) {
+    clip_bias_us_ += time_us - system_time_us;
+    time_us = system_time_us;
+  }
+  prev_translated_time_us_ = rtc::Optional<int64_t>(time_us);
+  return time_us;
+}
+
 bool VideoCapturer::AdaptFrame(int width,
                                int height,
-                               // TODO(nisse): Switch to us unit.
-                               int64_t capture_time_ns,
+                               int64_t camera_time_us,
+                               int64_t system_time_us,
                                int* out_width,
                                int* out_height,
                                int* crop_width,
                                int* crop_height,
                                int* crop_x,
-                               int* crop_y) {
+                               int* crop_y,
+                               int64_t* translated_camera_time_us) {
+  if (translated_camera_time_us) {
+    UpdateOffset(camera_time_us, system_time_us);
+  }
   if (!broadcaster_.frame_wanted()) {
     return false;
   }
 
   if (enable_video_adapter_ && !IsScreencast()) {
     if (!video_adapter_.AdaptFrameResolution(
-            width, height, capture_time_ns,
+            width, height, camera_time_us * rtc::kNumNanosecsPerMicrosec,
             crop_width, crop_height, out_width, out_height)) {
       // VideoAdapter dropped the frame.
       return false;
@@ -245,6 +336,11 @@ bool VideoCapturer::AdaptFrame(int width,
     *crop_x = 0;
     *crop_y = 0;
   }
+
+  if (translated_camera_time_us) {
+    *translated_camera_time_us =
+        ClipTimestamp(camera_time_us + offset_us_, system_time_us);
+  }
   return true;
 }
 
@@ -257,10 +353,17 @@ void VideoCapturer::OnFrameCaptured(VideoCapturer*,
   int crop_x;
   int crop_y;
 
+  // TODO(nisse): We don't do timestamp translation on this input
+  // path. It seems straight-forward to enable translation, but that
+  // breaks the WebRtcVideoEngine2Test.PropagatesInputFrameTimestamp
+  // test. Probably not worth the effort to fix, instead, try to
+  // delete or refactor all code using VideoFrameFactory and
+  // SignalCapturedFrame.
   if (!AdaptFrame(captured_frame->width, captured_frame->height,
-                  captured_frame->time_stamp,
+                  captured_frame->time_stamp / rtc::kNumNanosecsPerMicrosec,
+                  0,
                   &out_width, &out_height,
-                  &crop_width, &crop_height, &crop_x, &crop_y)) {
+                  &crop_width, &crop_height, &crop_x, &crop_y, nullptr)) {
     return;
   }