vpx/test/test_vector_test.cc

177 lines
5.6 KiB
C++
Raw Normal View History

/*
* Copyright (c) 2013 The WebM 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.
*/
#include <cstdio>
#include <cstdlib>
#include <string>
#include "third_party/googletest/src/include/gtest/gtest.h"
#include "../tools_common.h"
#include "./vpx_config.h"
#include "test/codec_factory.h"
#include "test/decode_test_driver.h"
#include "test/ivf_video_source.h"
[WIP] Add column-based tiling. This patch adds column-based tiling. The idea is to make each tile independently decodable (after reading the common frame header) and also independendly encodable (minus within-frame cost adjustments in the RD loop) to speed-up hardware & software en/decoders if they used multi-threading. Column-based tiling has the added advantage (over other tiling methods) that it minimizes realtime use-case latency, since all threads can start encoding data as soon as the first SB-row worth of data is available to the encoder. There is some test code that does random tile ordering in the decoder, to confirm that each tile is indeed independently decodable from other tiles in the same frame. At tile edges, all contexts assume default values (i.e. 0, 0 motion vector, no coefficients, DC intra4x4 mode), and motion vector search and ordering do not cross tiles in the same frame. t log Tile independence is not maintained between frames ATM, i.e. tile 0 of frame 1 is free to use motion vectors that point into any tile of frame 0. We support 1 (i.e. no tiling), 2 or 4 column-tiles. The loopfilter crosses tile boundaries. I discussed this briefly with Aki and he says that's OK. An in-loop loopfilter would need to do some sync between tile threads, but that shouldn't be a big issue. Resuls: with tiling disabled, we go up slightly because of improved edge use in the intra4x4 prediction. With 2 tiles, we lose about ~1% on derf, ~0.35% on HD and ~0.55% on STD/HD. With 4 tiles, we lose another ~1.5% on derf ~0.77% on HD and ~0.85% on STD/HD. Most of this loss is concentrated in the low-bitrate end of clips, and most of it is because of the loss of edges at tile boundaries and the resulting loss of intra predictors. TODO: - more tiles (perhaps allow row-based tiling also, and max. 8 tiles)? - maybe optionally (for EC purposes), motion vectors themselves should not cross tile edges, or we should emulate such borders as if they were off-frame, to limit error propagation to within one tile only. This doesn't have to be the default behaviour but could be an optional bitstream flag. Change-Id: I5951c3a0742a767b20bc9fb5af685d9892c2c96f
2013-02-01 18:35:28 +01:00
#include "test/md5_helper.h"
#include "test/test_vectors.h"
#include "test/util.h"
#if CONFIG_WEBM_IO
#include "test/webm_video_source.h"
#endif
#include "vpx_mem/vpx_mem.h"
namespace {
enum DecodeMode {
kSerialMode,
kFrameParallMode
};
const int kDecodeMode = 0;
const int kThreads = 1;
const int kFileName = 2;
typedef std::tr1::tuple<int, int, const char *> DecodeParam;
class TestVectorTest : public ::libvpx_test::DecoderTest,
public ::libvpx_test::CodecTestWithParam<DecodeParam> {
protected:
TestVectorTest()
: DecoderTest(GET_PARAM(0)),
md5_file_(NULL) {
}
virtual ~TestVectorTest() {
if (md5_file_)
fclose(md5_file_);
}
void OpenMD5File(const std::string& md5_file_name_) {
md5_file_ = libvpx_test::OpenTestDataFile(md5_file_name_);
ASSERT_TRUE(md5_file_ != NULL) << "Md5 file open failed. Filename: "
<< md5_file_name_;
}
virtual void DecompressedFrameHook(const vpx_image_t& img,
const unsigned int frame_number) {
ASSERT_TRUE(md5_file_ != NULL);
char expected_md5[33];
char junk[128];
// Read correct md5 checksums.
const int res = fscanf(md5_file_, "%s %s", expected_md5, junk);
ASSERT_NE(res, EOF) << "Read md5 data failed";
expected_md5[32] = '\0';
[WIP] Add column-based tiling. This patch adds column-based tiling. The idea is to make each tile independently decodable (after reading the common frame header) and also independendly encodable (minus within-frame cost adjustments in the RD loop) to speed-up hardware & software en/decoders if they used multi-threading. Column-based tiling has the added advantage (over other tiling methods) that it minimizes realtime use-case latency, since all threads can start encoding data as soon as the first SB-row worth of data is available to the encoder. There is some test code that does random tile ordering in the decoder, to confirm that each tile is indeed independently decodable from other tiles in the same frame. At tile edges, all contexts assume default values (i.e. 0, 0 motion vector, no coefficients, DC intra4x4 mode), and motion vector search and ordering do not cross tiles in the same frame. t log Tile independence is not maintained between frames ATM, i.e. tile 0 of frame 1 is free to use motion vectors that point into any tile of frame 0. We support 1 (i.e. no tiling), 2 or 4 column-tiles. The loopfilter crosses tile boundaries. I discussed this briefly with Aki and he says that's OK. An in-loop loopfilter would need to do some sync between tile threads, but that shouldn't be a big issue. Resuls: with tiling disabled, we go up slightly because of improved edge use in the intra4x4 prediction. With 2 tiles, we lose about ~1% on derf, ~0.35% on HD and ~0.55% on STD/HD. With 4 tiles, we lose another ~1.5% on derf ~0.77% on HD and ~0.85% on STD/HD. Most of this loss is concentrated in the low-bitrate end of clips, and most of it is because of the loss of edges at tile boundaries and the resulting loss of intra predictors. TODO: - more tiles (perhaps allow row-based tiling also, and max. 8 tiles)? - maybe optionally (for EC purposes), motion vectors themselves should not cross tile edges, or we should emulate such borders as if they were off-frame, to limit error propagation to within one tile only. This doesn't have to be the default behaviour but could be an optional bitstream flag. Change-Id: I5951c3a0742a767b20bc9fb5af685d9892c2c96f
2013-02-01 18:35:28 +01:00
::libvpx_test::MD5 md5_res;
md5_res.Add(&img);
const char *actual_md5 = md5_res.Get();
// Check md5 match.
ASSERT_STREQ(expected_md5, actual_md5)
<< "Md5 checksums don't match: frame number = " << frame_number;
}
private:
FILE *md5_file_;
};
// This test runs through the whole set of test vectors, and decodes them.
// The md5 checksums are computed for each frame in the video file. If md5
// checksums match the correct md5 data, then the test is passed. Otherwise,
// the test failed.
TEST_P(TestVectorTest, MD5Match) {
const DecodeParam input = GET_PARAM(1);
const std::string filename = std::tr1::get<kFileName>(input);
const int threads = std::tr1::get<kThreads>(input);
const int mode = std::tr1::get<kDecodeMode>(input);
libvpx_test::CompressedVideoSource *video = NULL;
vpx_codec_flags_t flags = 0;
vpx_codec_dec_cfg_t cfg = {0};
char str[256];
if (mode == kFrameParallMode) {
flags |= VPX_CODEC_USE_FRAME_THREADING;
}
cfg.threads = threads;
snprintf(str, sizeof(str) / sizeof(str[0]) - 1,
"file: %s mode: %s threads: %d",
filename.c_str(), mode == 0 ? "Serial" : "Parallel", threads);
SCOPED_TRACE(str);
// Open compressed video file.
if (filename.substr(filename.length() - 3, 3) == "ivf") {
video = new libvpx_test::IVFVideoSource(filename);
} else if (filename.substr(filename.length() - 4, 4) == "webm") {
#if CONFIG_WEBM_IO
video = new libvpx_test::WebMVideoSource(filename);
#else
fprintf(stderr, "WebM IO is disabled, skipping test vector %s\n",
filename.c_str());
return;
#endif
}
video->Init();
// Construct md5 file name.
const std::string md5_filename = filename + ".md5";
OpenMD5File(md5_filename);
// Set decode config and flags.
set_cfg(cfg);
set_flags(flags);
// Decode frame, and check the md5 matching.
ASSERT_NO_FATAL_FAILURE(RunLoop(video));
delete video;
}
// Test VP8 decode in serial mode with single thread.
// NOTE: VP8 only support serial mode.
INSTANTIATE_TEST_CASE_P(
VP8, TestVectorTest,
::testing::Combine(
::testing::Values(
static_cast<const libvpx_test::CodecFactory *>(&libvpx_test::kVP8)),
::testing::Combine(
::testing::Values(0), // Serial Mode.
::testing::Values(1), // Single thread.
::testing::ValuesIn(libvpx_test::kVP8TestVectors,
libvpx_test::kVP8TestVectors +
libvpx_test::kNumVP8TestVectors))));
// Test VP9 decode in serial mode with single thread.
INSTANTIATE_TEST_CASE_P(
VP9, TestVectorTest,
::testing::Combine(
::testing::Values(
static_cast<const libvpx_test::CodecFactory *>(&libvpx_test::kVP9)),
::testing::Combine(
::testing::Values(0), // Serial Mode.
::testing::Values(1), // Single thread.
::testing::ValuesIn(libvpx_test::kVP9TestVectors,
libvpx_test::kVP9TestVectors +
libvpx_test::kNumVP9TestVectors))));
// Test VP9 decode in frame parallel mode with different number of threads.
INSTANTIATE_TEST_CASE_P(
VP9MultiThreadedFrameParallel, TestVectorTest,
::testing::Combine(
::testing::Values(
static_cast<const libvpx_test::CodecFactory *>(&libvpx_test::kVP9)),
::testing::Combine(
::testing::Values(1), // Frame Parallel mode.
::testing::Values(2, 3, 4), // With 2, 3, 4 threads.
::testing::ValuesIn(libvpx_test::kVP9TestVectors,
libvpx_test::kVP9TestVectors +
libvpx_test::kNumVP9TestVectors))));
} // namespace