1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
|
/*
* Copyright (c) 2020 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.
*/
#include <stddef.h>
#include <stdint.h>
#include <memory>
#include <optional>
#include <ostream>
#include <set>
#include <string>
#include <utility>
#include <vector>
#include "api/array_view.h"
#include "api/transport/rtp/dependency_descriptor.h"
#include "api/video/video_bitrate_allocation.h"
#include "api/video_codecs/scalability_mode.h"
#include "common_video/generic_frame_descriptor/generic_frame_info.h"
#include "modules/video_coding/svc/create_scalability_structure.h"
#include "modules/video_coding/svc/scalability_mode_util.h"
#include "modules/video_coding/svc/scalability_structure_test_helpers.h"
#include "modules/video_coding/svc/scalable_video_controller.h"
#include "rtc_base/checks.h"
#include "rtc_base/strings/string_builder.h"
#include "test/gmock.h"
#include "test/gtest.h"
namespace webrtc {
namespace {
using ::testing::AllOf;
using ::testing::Contains;
using ::testing::Each;
using ::testing::ElementsAreArray;
using ::testing::Field;
using ::testing::Ge;
using ::testing::IsEmpty;
using ::testing::Le;
using ::testing::Lt;
using ::testing::Not;
using ::testing::NotNull;
using ::testing::SizeIs;
using ::testing::TestWithParam;
using ::testing::Values;
std::string FrameDependencyTemplateToString(const FrameDependencyTemplate& t) {
StringBuilder sb;
sb << "S" << t.spatial_id << "T" << t.temporal_id;
sb << ": dtis = ";
for (const auto dtis : t.decode_target_indications) {
switch (dtis) {
case DecodeTargetIndication::kNotPresent:
sb << "-";
break;
case DecodeTargetIndication::kDiscardable:
sb << "D";
break;
case DecodeTargetIndication::kSwitch:
sb << "S";
break;
case DecodeTargetIndication::kRequired:
sb << "R";
break;
default:
sb << "?";
break;
}
}
sb << ", frame diffs = { ";
for (int d : t.frame_diffs) {
sb << d << ", ";
}
sb << "}, chain diffs = { ";
for (int d : t.chain_diffs) {
sb << d << ", ";
}
sb << "}";
return sb.Release();
}
struct SvcTestParam {
friend std::ostream& operator<<(std::ostream& os, const SvcTestParam& param) {
return os << param.name;
}
ScalabilityMode GetScalabilityMode() const {
std::optional<ScalabilityMode> scalability_mode =
ScalabilityModeFromString(name);
RTC_CHECK(scalability_mode.has_value());
return *scalability_mode;
}
std::string name;
int num_temporal_units;
};
class ScalabilityStructureTest : public TestWithParam<SvcTestParam> {};
TEST_P(ScalabilityStructureTest,
StaticConfigMatchesConfigReturnedByController) {
std::unique_ptr<ScalableVideoController> controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
std::optional<ScalableVideoController::StreamLayersConfig> static_config =
ScalabilityStructureConfig(GetParam().GetScalabilityMode());
ASSERT_THAT(controller, NotNull());
ASSERT_NE(static_config, std::nullopt);
ScalableVideoController::StreamLayersConfig config =
controller->StreamConfig();
EXPECT_EQ(config.num_spatial_layers, static_config->num_spatial_layers);
EXPECT_EQ(config.num_temporal_layers, static_config->num_temporal_layers);
EXPECT_THAT(
MakeArrayView(config.scaling_factor_num, config.num_spatial_layers),
ElementsAreArray(static_config->scaling_factor_num,
static_config->num_spatial_layers));
EXPECT_THAT(
MakeArrayView(config.scaling_factor_den, config.num_spatial_layers),
ElementsAreArray(static_config->scaling_factor_den,
static_config->num_spatial_layers));
}
TEST_P(ScalabilityStructureTest,
NumberOfDecodeTargetsAndChainsAreInRangeAndConsistent) {
FrameDependencyStructure structure =
CreateScalabilityStructure(GetParam().GetScalabilityMode())
->DependencyStructure();
EXPECT_GT(structure.num_decode_targets, 0);
EXPECT_LE(structure.num_decode_targets,
DependencyDescriptor::kMaxDecodeTargets);
EXPECT_GE(structure.num_chains, 0);
EXPECT_LE(structure.num_chains, structure.num_decode_targets);
if (structure.num_chains == 0) {
EXPECT_THAT(structure.decode_target_protected_by_chain, IsEmpty());
} else {
EXPECT_THAT(structure.decode_target_protected_by_chain,
AllOf(SizeIs(structure.num_decode_targets), Each(Ge(0)),
Each(Lt(structure.num_chains))));
}
EXPECT_THAT(structure.templates,
SizeIs(Lt(size_t{DependencyDescriptor::kMaxTemplates})));
}
TEST_P(ScalabilityStructureTest, TemplatesAreSortedByLayerId) {
FrameDependencyStructure structure =
CreateScalabilityStructure(GetParam().GetScalabilityMode())
->DependencyStructure();
ASSERT_THAT(structure.templates, Not(IsEmpty()));
const auto& first_templates = structure.templates.front();
EXPECT_EQ(first_templates.spatial_id, 0);
EXPECT_EQ(first_templates.temporal_id, 0);
for (size_t i = 1; i < structure.templates.size(); ++i) {
const auto& prev_template = structure.templates[i - 1];
const auto& next_template = structure.templates[i];
if (next_template.spatial_id == prev_template.spatial_id &&
next_template.temporal_id == prev_template.temporal_id) {
// Same layer, next_layer_idc == 0
} else if (next_template.spatial_id == prev_template.spatial_id &&
next_template.temporal_id == prev_template.temporal_id + 1) {
// Next temporal layer, next_layer_idc == 1
} else if (next_template.spatial_id == prev_template.spatial_id + 1 &&
next_template.temporal_id == 0) {
// Next spatial layer, next_layer_idc == 2
} else {
// everything else is invalid.
ADD_FAILURE() << "Invalid templates order. Template #" << i
<< " with layer (" << next_template.spatial_id << ","
<< next_template.temporal_id
<< ") follows template with layer ("
<< prev_template.spatial_id << ","
<< prev_template.temporal_id << ").";
}
}
}
TEST_P(ScalabilityStructureTest, TemplatesMatchNumberOfDecodeTargetsAndChains) {
FrameDependencyStructure structure =
CreateScalabilityStructure(GetParam().GetScalabilityMode())
->DependencyStructure();
EXPECT_THAT(
structure.templates,
Each(AllOf(Field(&FrameDependencyTemplate::decode_target_indications,
SizeIs(structure.num_decode_targets)),
Field(&FrameDependencyTemplate::chain_diffs,
SizeIs(structure.num_chains)))));
}
TEST_P(ScalabilityStructureTest, FrameInfoMatchesFrameDependencyStructure) {
std::unique_ptr<ScalableVideoController> svc_controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
FrameDependencyStructure structure = svc_controller->DependencyStructure();
std::vector<GenericFrameInfo> frame_infos =
ScalabilityStructureWrapper(*svc_controller)
.GenerateFrames(GetParam().num_temporal_units);
for (size_t frame_id = 0; frame_id < frame_infos.size(); ++frame_id) {
const auto& frame = frame_infos[frame_id];
EXPECT_GE(frame.spatial_id, 0) << " for frame " << frame_id;
EXPECT_GE(frame.temporal_id, 0) << " for frame " << frame_id;
EXPECT_THAT(frame.decode_target_indications,
SizeIs(structure.num_decode_targets))
<< " for frame " << frame_id;
EXPECT_THAT(frame.part_of_chain, SizeIs(structure.num_chains))
<< " for frame " << frame_id;
}
}
TEST_P(ScalabilityStructureTest, ThereIsAPerfectTemplateForEachFrame) {
std::unique_ptr<ScalableVideoController> svc_controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
FrameDependencyStructure structure = svc_controller->DependencyStructure();
std::vector<GenericFrameInfo> frame_infos =
ScalabilityStructureWrapper(*svc_controller)
.GenerateFrames(GetParam().num_temporal_units);
for (size_t frame_id = 0; frame_id < frame_infos.size(); ++frame_id) {
EXPECT_THAT(structure.templates, Contains(frame_infos[frame_id]))
<< " for frame " << frame_id << ", Expected "
<< FrameDependencyTemplateToString(frame_infos[frame_id]);
}
}
TEST_P(ScalabilityStructureTest, FrameDependsOnSameOrLowerLayer) {
std::unique_ptr<ScalableVideoController> svc_controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
std::vector<GenericFrameInfo> frame_infos =
ScalabilityStructureWrapper(*svc_controller)
.GenerateFrames(GetParam().num_temporal_units);
int64_t num_frames = frame_infos.size();
for (int64_t frame_id = 0; frame_id < num_frames; ++frame_id) {
const auto& frame = frame_infos[frame_id];
for (int frame_diff : frame.frame_diffs) {
int64_t base_frame_id = frame_id - frame_diff;
const auto& base_frame = frame_infos[base_frame_id];
EXPECT_GE(frame.spatial_id, base_frame.spatial_id)
<< "Frame " << frame_id << " depends on frame " << base_frame_id;
EXPECT_GE(frame.temporal_id, base_frame.temporal_id)
<< "Frame " << frame_id << " depends on frame " << base_frame_id;
}
}
}
TEST_P(ScalabilityStructureTest, NoFrameDependsOnDiscardableOrNotPresent) {
std::unique_ptr<ScalableVideoController> svc_controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
std::vector<GenericFrameInfo> frame_infos =
ScalabilityStructureWrapper(*svc_controller)
.GenerateFrames(GetParam().num_temporal_units);
int64_t num_frames = frame_infos.size();
FrameDependencyStructure structure = svc_controller->DependencyStructure();
for (int dt = 0; dt < structure.num_decode_targets; ++dt) {
for (int64_t frame_id = 0; frame_id < num_frames; ++frame_id) {
const auto& frame = frame_infos[frame_id];
if (frame.decode_target_indications[dt] ==
DecodeTargetIndication::kNotPresent) {
continue;
}
for (int frame_diff : frame.frame_diffs) {
int64_t base_frame_id = frame_id - frame_diff;
const auto& base_frame = frame_infos[base_frame_id];
EXPECT_NE(base_frame.decode_target_indications[dt],
DecodeTargetIndication::kNotPresent)
<< "Frame " << frame_id << " depends on frame " << base_frame_id
<< " that is not part of decode target#" << dt;
EXPECT_NE(base_frame.decode_target_indications[dt],
DecodeTargetIndication::kDiscardable)
<< "Frame " << frame_id << " depends on frame " << base_frame_id
<< " that is discardable for decode target#" << dt;
}
}
}
}
TEST_P(ScalabilityStructureTest, NoFrameDependsThroughSwitchIndication) {
std::unique_ptr<ScalableVideoController> svc_controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
FrameDependencyStructure structure = svc_controller->DependencyStructure();
std::vector<GenericFrameInfo> frame_infos =
ScalabilityStructureWrapper(*svc_controller)
.GenerateFrames(GetParam().num_temporal_units);
int64_t num_frames = frame_infos.size();
std::vector<std::set<int64_t>> full_deps(num_frames);
// For each frame calculate set of all frames it depends on, both directly and
// indirectly.
for (int64_t frame_id = 0; frame_id < num_frames; ++frame_id) {
std::set<int64_t> all_base_frames;
for (int frame_diff : frame_infos[frame_id].frame_diffs) {
int64_t base_frame_id = frame_id - frame_diff;
all_base_frames.insert(base_frame_id);
const auto& indirect = full_deps[base_frame_id];
all_base_frames.insert(indirect.begin(), indirect.end());
}
full_deps[frame_id] = std::move(all_base_frames);
}
// Now check the switch indication: frames after the switch indication mustn't
// depend on any addition frames before the switch indications.
for (int dt = 0; dt < structure.num_decode_targets; ++dt) {
for (int64_t switch_frame_id = 0; switch_frame_id < num_frames;
++switch_frame_id) {
if (frame_infos[switch_frame_id].decode_target_indications[dt] !=
DecodeTargetIndication::kSwitch) {
continue;
}
for (int64_t later_frame_id = switch_frame_id + 1;
later_frame_id < num_frames; ++later_frame_id) {
if (frame_infos[later_frame_id].decode_target_indications[dt] ==
DecodeTargetIndication::kNotPresent) {
continue;
}
for (int frame_diff : frame_infos[later_frame_id].frame_diffs) {
int64_t early_frame_id = later_frame_id - frame_diff;
if (early_frame_id < switch_frame_id) {
EXPECT_THAT(full_deps[switch_frame_id], Contains(early_frame_id))
<< "For decode target #" << dt << " frame " << later_frame_id
<< " depends on the frame " << early_frame_id
<< " that switch indication frame " << switch_frame_id
<< " doesn't directly on indirectly depend on.";
}
}
}
}
}
}
TEST_P(ScalabilityStructureTest, ProduceNoFrameForDisabledLayers) {
std::unique_ptr<ScalableVideoController> svc_controller =
CreateScalabilityStructure(GetParam().GetScalabilityMode());
ScalableVideoController::StreamLayersConfig structure =
svc_controller->StreamConfig();
VideoBitrateAllocation all_bitrates;
for (int sid = 0; sid < structure.num_spatial_layers; ++sid) {
for (int tid = 0; tid < structure.num_temporal_layers; ++tid) {
all_bitrates.SetBitrate(sid, tid, 100'000);
}
}
svc_controller->OnRatesUpdated(all_bitrates);
ScalabilityStructureWrapper wrapper(*svc_controller);
std::vector<GenericFrameInfo> frames =
wrapper.GenerateFrames(GetParam().num_temporal_units);
for (int sid = 0; sid < structure.num_spatial_layers; ++sid) {
for (int tid = 0; tid < structure.num_temporal_layers; ++tid) {
// When all layers were enabled, expect there was a frame for each layer.
EXPECT_THAT(frames,
Contains(AllOf(Field(&GenericFrameInfo::spatial_id, sid),
Field(&GenericFrameInfo::temporal_id, tid))))
<< "For layer (" << sid << "," << tid << ")";
// Restore bitrates for all layers before disabling single layer.
VideoBitrateAllocation bitrates = all_bitrates;
bitrates.SetBitrate(sid, tid, 0);
svc_controller->OnRatesUpdated(bitrates);
// With layer (sid, tid) disabled, expect no frames are produced for it.
EXPECT_THAT(
wrapper.GenerateFrames(GetParam().num_temporal_units),
Not(Contains(AllOf(Field(&GenericFrameInfo::spatial_id, sid),
Field(&GenericFrameInfo::temporal_id, tid)))))
<< "For layer (" << sid << "," << tid << ")";
}
}
}
INSTANTIATE_TEST_SUITE_P(
Svc,
ScalabilityStructureTest,
Values(SvcTestParam{"L1T1", /*num_temporal_units=*/3},
SvcTestParam{"L1T2", /*num_temporal_units=*/4},
SvcTestParam{"L1T3", /*num_temporal_units=*/8},
SvcTestParam{"L2T1", /*num_temporal_units=*/3},
SvcTestParam{"L2T1_KEY", /*num_temporal_units=*/3},
SvcTestParam{"L3T1", /*num_temporal_units=*/3},
SvcTestParam{"L3T1_KEY", /*num_temporal_units=*/3},
SvcTestParam{"L3T3", /*num_temporal_units=*/8},
SvcTestParam{"S2T1", /*num_temporal_units=*/3},
SvcTestParam{"S2T2", /*num_temporal_units=*/4},
SvcTestParam{"S2T3", /*num_temporal_units=*/8},
SvcTestParam{"S3T1", /*num_temporal_units=*/3},
SvcTestParam{"S3T2", /*num_temporal_units=*/4},
SvcTestParam{"S3T3", /*num_temporal_units=*/8},
SvcTestParam{"L2T2", /*num_temporal_units=*/4},
SvcTestParam{"L2T2_KEY", /*num_temporal_units=*/4},
SvcTestParam{"L2T2_KEY_SHIFT", /*num_temporal_units=*/4},
SvcTestParam{"L2T3", /*num_temporal_units=*/8},
SvcTestParam{"L2T3_KEY", /*num_temporal_units=*/8},
SvcTestParam{"L3T2", /*num_temporal_units=*/4},
SvcTestParam{"L3T2_KEY", /*num_temporal_units=*/4},
SvcTestParam{"L3T3_KEY", /*num_temporal_units=*/8}),
[](const testing::TestParamInfo<SvcTestParam>& info) {
return info.param.name;
});
} // namespace
} // namespace webrtc
|