This makes it easier to add new test cases without modifying the actual test class.
Bug: None
Change-Id: I48e4f14e26cd6610678ffb07ce9fd56e6bc1ac4e
Reviewed-on: https://webrtc-review.googlesource.com/69600
Commit-Queue: Kári Helgason <kthelgason@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22840}
* Add support for SimulcastEncoderAdapter wrapping of encoder.
* Store input frame timestamps out-of-band, so we don't need to keep
a raw VideoFrame around just for it's timestamp.
* Store current frame rate in |framerate_fps_|, instead of in
codec settings struct.
* Add some comments and reorder some data members.
* Explicitly include VideoBitrateAllocator.
* Change type of |input_frames_|, to avoid one layer of indirection.
* Move VideoProcessor::CalculateFrameQuality to anonymous namespace.
This change should have no functional implications.
Bug: webrtc:8448
Change-Id: I10c140eeda750d9bd37bfb6cb1e8acb401fb91d3
Reviewed-on: https://webrtc-review.googlesource.com/60520
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22346}
Previously, the only user of this code was the
VideoProcessorIntegrationTest. We have now changed that
test to directly calculate image quality metrics using libyuv,
similar to how the full stack tests and browser tests work.
Bug: webrtc:8448
Change-Id: Ia7a607d7ddc37741fba76d56aa7297851ffa1c6b
Reviewed-on: https://webrtc-review.googlesource.com/43760
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22341}
* Do not simulate freeze in decoded output file when frames have been dropped.
* Add more DCHECKs and consts.
* Remove unused members |num_encoded_frames_| and |num_decoded_frames_|.
* Move SdpVideoFormat conversion to TestConfig.
Bug: webrtc:8448
Change-Id: Ia879141f36dc23427cd1abcaa66716656fbaac2a
Reviewed-on: https://webrtc-review.googlesource.com/56802
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22239}
Slicing, aggregation and analysis has been moved to Stats class.
Data of all spatial layers is stored in single Stats object.
Bug: webrtc:8524
Change-Id: Ic9a64859a36a1ccda661942a201cdeeed470686a
Reviewed-on: https://webrtc-review.googlesource.com/50301
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22094}
Accessing this method from the test thread is illegal,
but doesn't always fail.
Bug: webrtc:8524
Change-Id: Ie0e84cc2fb63268fb6d7cbf0c3a58cb35312c16b
Reviewed-on: https://webrtc-review.googlesource.com/49061
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21930}
This helps separate concerns, so that the VideoProcessorIntegrationTest
is almost oblivious to the fact that it needs to connect to the JVM
to get the Android HW codecs.
Bug: webrtc:8448
Change-Id: I4359b31f84be48eaf99d83525bcce6e593e874f8
Reviewed-on: https://webrtc-review.googlesource.com/47384
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21890}
Encoded frames are preserved and decoded after all layers are
encoded.
Each spatial layer is decoded with separate decoder.
For quality evaluation of lowres layers original input frame is
downscaled with bilinear interpolation.
Encoded and decoded frames are dumped into separate files.
For async codecs encoded frames are passed to decoder in encode
callback, as before.
Bug: webrtc:8524
Change-Id: Idb0c92c7274c1915cff9a011a2794f1cf4bc8cb1
Reviewed-on: https://webrtc-review.googlesource.com/43381
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21844}
Each simulcast stream requires dedicated decoder for decoding. SVC
can be decoded by single decoder. But in prod each receiver has its
decoder. We want to replicate this and also use one decoder per
spatial layer.
Also we create one frame writer per simulcast/spatial layer to dump
encoded/decoded frames of different layers to separate files.
Note that videoprocessor is still initialized with single
decoder/writer. It will be updated in next CL and start using
separate decoder/writer per layer.
Bug: webrtc:8524
Change-Id: I3bb3de77f97d51138b8b7675dd01bc281a078b2f
Reviewed-on: https://webrtc-review.googlesource.com/43280
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21744}
This feature is not needed in video codec testing framework. In WebRTC
video codecs never deal with packet loss. Packet loss is handled by
jitter buffer which prevents passing of incomplete frames to decoder.
Bug: webrtc:8768
Change-Id: I211cf51d913bec6a1f935e30691661d428ebd3b6
Reviewed-on: https://webrtc-review.googlesource.com/40740
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21722}
This is a reland of 1880c7162b
Original change's description:
> Updated analysis in videoprocessor.
>
> - Run analysis after all frames are processed. Before part of it was
> done at bitrate change points;
> - Analysis is done for whole stream as well as for each rate update
> interval;
> - Changed units from number of frames to time units for some metrics
> and thresholds. E.g. 'num frames to hit tagret bitrate' is changed to
> 'time to reach target bitrate, sec';
> - Changed data type of FrameStatistic::max_nalu_length (renamed to
> max_nalu_size_bytes) from rtc::Optional to size_t. There it no need to
> use such advanced data type in such low level data structure.
>
> Bug: webrtc:8524
> Change-Id: Ic9f6eab5b15ee12a80324b1f9c101de1bf3c702f
> Reviewed-on: https://webrtc-review.googlesource.com/31901
> Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Åsa Persson <asapersson@webrtc.org>
> Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#21653}
TBR=brandtr@webrtc.org, stefan@webrtc.org
Bug: webrtc:8524
Change-Id: Ie0ad7790689422ffa61da294967fc492a13b75ae
Reviewed-on: https://webrtc-review.googlesource.com/40202
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21668}
This reverts commit 1880c7162b.
Reason for revert: breaks internal tests
Original change's description:
> Updated analysis in videoprocessor.
>
> - Run analysis after all frames are processed. Before part of it was
> done at bitrate change points;
> - Analysis is done for whole stream as well as for each rate update
> interval;
> - Changed units from number of frames to time units for some metrics
> and thresholds. E.g. 'num frames to hit tagret bitrate' is changed to
> 'time to reach target bitrate, sec';
> - Changed data type of FrameStatistic::max_nalu_length (renamed to
> max_nalu_size_bytes) from rtc::Optional to size_t. There it no need to
> use such advanced data type in such low level data structure.
>
> Bug: webrtc:8524
> Change-Id: Ic9f6eab5b15ee12a80324b1f9c101de1bf3c702f
> Reviewed-on: https://webrtc-review.googlesource.com/31901
> Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Åsa Persson <asapersson@webrtc.org>
> Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#21653}
TBR=brandtr@webrtc.org,asapersson@webrtc.org,sprang@webrtc.org,stefan@webrtc.org,ssilkin@webrtc.org
Change-Id: Id0b7d387bbba02e71637b229aeed6f6cf012af46
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:8524
Reviewed-on: https://webrtc-review.googlesource.com/40220
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21656}
- Run analysis after all frames are processed. Before part of it was
done at bitrate change points;
- Analysis is done for whole stream as well as for each rate update
interval;
- Changed units from number of frames to time units for some metrics
and thresholds. E.g. 'num frames to hit tagret bitrate' is changed to
'time to reach target bitrate, sec';
- Changed data type of FrameStatistic::max_nalu_length (renamed to
max_nalu_size_bytes) from rtc::Optional to size_t. There it no need to
use such advanced data type in such low level data structure.
Bug: webrtc:8524
Change-Id: Ic9f6eab5b15ee12a80324b1f9c101de1bf3c702f
Reviewed-on: https://webrtc-review.googlesource.com/31901
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21653}
We currently use raw jobject in our code mixed with sporadic
ScopedLocalRefFrame. This CL moves every jobject into a scoped object,
either local, global, or a parameter. Also, this CL uses the JNI
generation script to generate declaration stubs for the Java->C++
functions so that it no longer becomes possible to mistype them
without getting compilation errors.
TBR=brandt@webrtc.org
Bug: webrtc:8278,webrtc:6969
Change-Id: Ic7bac74a89c11180177d65041086d7db1cdfb516
Reviewed-on: https://webrtc-review.googlesource.com/34655
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Reviewed-by: Sami Kalliomäki <sakal@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21387}
This will also cause us to use the new Android HardwareVideoEncoder,
instead of the deprecated MediaCodecVideoEncoderFactory. Unfortunately,
the new HW encoder does not seem to work as good as the old (or the new
encoder is more strict with return values or something). I don't think
it adds much value to continue testing the deprecated encoder, so I
filed a bug for fixing the new encoder, and in this CL I disabled the
tests on Android. I want to remove as many places as possible where we
use the old WebRtcVideoEncoderFactory interface, because it makes it
more difficult to migrate to the new interface.
Bug: webrtc:7925
Change-Id: If8e34752148a5e5139944d2dfbe7e231fe58aeb9
Reviewed-on: https://webrtc-review.googlesource.com/27540
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Reviewed-by: Sami Kalliomäki <sakal@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21037}
Calculation of quality metrics required writing of decoded video
to file. There were two drawbacks with that approach. First, frame
drops significantly affected metrics because comparison was done
against the last decoded frame. Second, simulcast/SVC required
writing of multiple files. This might be too much data to dump.
On-fly metrics calculation is done in frame decoded callback.
Calculation time is excluded from encoding/decoding time. If CPU
usage measurement is enabled metrics calculation is disabled since
it affects CPU usage. The results are reported in Stats::PrintSummary.
Bug: webrtc:8524
Change-Id: Id54fb21f2f95deeb93757afaf46bde7d7ae18dac
Reviewed-on: https://webrtc-review.googlesource.com/22560
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Patrik Höglund <phoglund@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20798}
VideoEncoderSoftwareFallbackWrapper is updated to take a VideoEncoder as
argument instead relying on built-in SW codecs. The purpose is to make
VideoEncoderSoftwareFallbackWrapper more modular and not depend on
built-in SW encoders.
Bug: webrtc:7925
Change-Id: I99896f0751cfb77e01efd29c97d3bd07bdb2c7c0
Reviewed-on: https://webrtc-review.googlesource.com/22320
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Anders Carlsson <andersc@webrtc.org>
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20671}
This reverts commit 267d84baf0.
Reason for reland: Fix the bug; decoder is not allowed to ever be null and we need to use a
NullVideoDecoder that ignores calls instead.
Original change's description:
> Revert "Update internal video decoder factory to new interface"
>
> This reverts commit b2fc9b1b10.
>
> Reason for revert: Suspected to cause failures on Android bots on webrtc.fyi, see https://build.chromium.org/p/chromium.webrtc.fyi/builders/Android%20Tests%20%28dbg%29%20%28K%20Nexus5%29/builds/21051
>
> Original change's description:
> > Update internal video decoder factory to new interface
> >
> > We want to move away from cricket::WebRtcVideoDecoderFactory and this CL
> > updates the internal factory. Also, VideoDecoderSoftwareFallbackWrapper
> > is updated to take a VideoDecoder as argument instead of a factory so it
> > can be used with external SW decoders.
> >
> > Bug: webrtc:7925
> > Change-Id: Ie6dc6c24f8610a2129620c6e2b42e3cebb2ddef7
> > Reviewed-on: https://webrtc-review.googlesource.com/7301
> > Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> > Reviewed-by: Anders Carlsson <andersc@webrtc.org>
> > Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#20597}
>
> TBR=brandtr@webrtc.org,magjed@webrtc.org,andersc@webrtc.org
>
> Change-Id: I0a12c98fdc30f00d58c85ee7e088f50160d39724
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: webrtc:7925
> Reviewed-on: https://webrtc-review.googlesource.com/21420
> Reviewed-by: Christian Fremerey <chfremer@webrtc.org>
> Commit-Queue: Christian Fremerey <chfremer@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#20605}
TBR=brandtr@webrtc.org,magjed@webrtc.org,andersc@webrtc.org,chfremer@webrtc.org,chfremer@google.com
Change-Id: I6cf5794dc3fadfa86809a94da80b69dbb4c56f52
No-Try: true
Bug: webrtc:7925
Reviewed-on: https://webrtc-review.googlesource.com/21541
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Reviewed-by: Magnus Jedvert <magjed@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20623}
This reverts commit b2fc9b1b10.
Reason for revert: Suspected to cause failures on Android bots on webrtc.fyi, see https://build.chromium.org/p/chromium.webrtc.fyi/builders/Android%20Tests%20%28dbg%29%20%28K%20Nexus5%29/builds/21051
Original change's description:
> Update internal video decoder factory to new interface
>
> We want to move away from cricket::WebRtcVideoDecoderFactory and this CL
> updates the internal factory. Also, VideoDecoderSoftwareFallbackWrapper
> is updated to take a VideoDecoder as argument instead of a factory so it
> can be used with external SW decoders.
>
> Bug: webrtc:7925
> Change-Id: Ie6dc6c24f8610a2129620c6e2b42e3cebb2ddef7
> Reviewed-on: https://webrtc-review.googlesource.com/7301
> Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> Reviewed-by: Anders Carlsson <andersc@webrtc.org>
> Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#20597}
TBR=brandtr@webrtc.org,magjed@webrtc.org,andersc@webrtc.org
Change-Id: I0a12c98fdc30f00d58c85ee7e088f50160d39724
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:7925
Reviewed-on: https://webrtc-review.googlesource.com/21420
Reviewed-by: Christian Fremerey <chfremer@webrtc.org>
Commit-Queue: Christian Fremerey <chfremer@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20605}
We want to move away from cricket::WebRtcVideoDecoderFactory and this CL
updates the internal factory. Also, VideoDecoderSoftwareFallbackWrapper
is updated to take a VideoDecoder as argument instead of a factory so it
can be used with external SW decoders.
Bug: webrtc:7925
Change-Id: Ie6dc6c24f8610a2129620c6e2b42e3cebb2ddef7
Reviewed-on: https://webrtc-review.googlesource.com/7301
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Anders Carlsson <andersc@webrtc.org>
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20597}
Always enabling verbose mode means about 100% more text is printed,
but this should not be a problem as the only time that we explicitly
look at the logs is when the bots are failing, or when we want to save
all output for plotting.
BUG=webrtc:8448
Change-Id: Ia5feab5220d047440d15cddb7d3fbca1c5a4aaf5
Reviewed-on: https://webrtc-review.googlesource.com/16140
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20461}
This CL adds an EncodedFrameChecker interface which can be used by users
of the VideoProcessor to inject customized per-frame checks to the
encoding/decoding pipeline. This currently has two uses:
- Verifying that the QP parser works correctly for VP8 and VP9, by comparing the
parsed QP to that produced by libvpx.
- Verifying that our H.264 encoders always produce SPS/PPS/IDR in tandem.
TESTED=Galaxy S8, Pixel 2 XL, iPhone 7.
BUG=webrtc:8423
Change-Id: Ic3e401546e239a9ffaf2ed2907689cebb1127805
Reviewed-on: https://webrtc-review.googlesource.com/14559
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20409}
Test enables single-nalu mode, sets limit for nalu lenght and verifies
that encoder follows that limit.
I found that QP jumps significantly when the mode is enabled. In result
encoder might produce 4kbyte and 0.4kbyte frames back-to-back. But it
seems that happens only to couple of frames in the beginning. This
caused test to fail with default RC thresholds. To bypass this I
increased frame size mismatch threshold from 20 to 30%. This should be
Ok considering single-nalu mode is rare.
BUG=webrtc:8070
Review-Url: https://codereview.webrtc.org/3014623002
Cr-Commit-Position: refs/heads/master@{#20023}
- Group member variables into two structs: target rates/actual rates.
- Split verify and print of rate control metrics into separate functions.
- Rename member variables.
BUG=webrtc:6634
Review-Url: https://codereview.webrtc.org/3009423002
Cr-Commit-Position: refs/heads/master@{#19925}
In https://webrtc-review.googlesource.com/c/src/+/1560 we moved WebRTC
from src/webrtc to src/ (in order to preserve an healthy git history).
This CL takes care of fixing header guards, #include paths, etc...
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
TBR=tommi@webrtc.org
Bug: chromium:611808
Change-Id: Iea91618212bee0af16aa3f05071eab8f93706578
Reviewed-on: https://webrtc-review.googlesource.com/1561
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#19846}
In order to eliminate the WebRTC Subtree mirror in Chromium,
WebRTC is moving the content of the src/webrtc directory up
to the src/ directory.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
TBR=tommi@webrtc.org
Bug: chromium:611808
Change-Id: Iac59c5b51b950f174119565bac87955a7994bc38
Reviewed-on: https://webrtc-review.googlesource.com/1560
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#19845}
2017-09-15 04:25:06 +00:00
Renamed from webrtc/modules/video_coding/codecs/test/videoprocessor_integrationtest.cc (Browse further)
