[x265] [PATCH] port emergency mode feature from x264
sagar at multicorewareinc.com
sagar at multicorewareinc.com
Mon Jul 27 14:15:49 CEST 2015
# HG changeset patch
# User Sagar Kotecha <sagar at multicorewareinc.com>
# Date 1437997463 -19800
# Mon Jul 27 17:14:23 2015 +0530
# Node ID b554b7cba38d66ade3feda62a40878f143737366
# Parent b015514a93868e2dcf431d2e94b147861076a376
port emergency mode feature from x264
The emergency mode activates in vbv mode if frame qp > QP_MAX_SPEC
diff -r b015514a9386 -r b554b7cba38d source/common/quant.cpp
--- a/source/common/quant.cpp Sun Jul 26 16:12:32 2015 +0530
+++ b/source/common/quant.cpp Mon Jul 27 17:14:23 2015 +0530
@@ -447,13 +447,13 @@
primitives.cu[sizeIdx].dct(m_fencShortBuf, m_fencDctCoeff, trSize);
}
- if (m_nr)
+ if (m_nr && m_nr->nrOffset)
{
/* denoise is not applied to intra residual, so DST can be ignored */
int cat = sizeIdx + 4 * !isLuma + 8 * !isIntra;
int numCoeff = 1 << (log2TrSize * 2);
- primitives.denoiseDct(m_resiDctCoeff, m_nr->residualSum[cat], m_nr->offsetDenoise[cat], numCoeff);
- m_nr->count[cat]++;
+ primitives.denoiseDct(m_resiDctCoeff, m_nr->nrResidualSum[cat], m_nr->nrOffset[cat], numCoeff);
+ m_nr->nrCount[cat]++;
}
}
diff -r b015514a9386 -r b554b7cba38d source/common/quant.h
--- a/source/common/quant.h Sun Jul 26 16:12:32 2015 +0530
+++ b/source/common/quant.h Mon Jul 27 17:14:23 2015 +0530
@@ -71,6 +71,14 @@
uint16_t offsetDenoise[MAX_NUM_TR_CATEGORIES][MAX_NUM_TR_COEFFS];
uint32_t residualSum[MAX_NUM_TR_CATEGORIES][MAX_NUM_TR_COEFFS];
uint32_t count[MAX_NUM_TR_CATEGORIES];
+
+ uint16_t offsetEmergency[QP_MAX_MAX - QP_MAX_SPEC][MAX_NUM_TR_CATEGORIES][MAX_NUM_TR_COEFFS];
+ uint32_t residualSumEmergency[MAX_NUM_TR_CATEGORIES][MAX_NUM_TR_COEFFS];
+ uint32_t countEmergency[MAX_NUM_TR_CATEGORIES];
+
+ ALIGN_VAR_16(uint16_t, (*nrOffset)[MAX_NUM_TR_COEFFS]);
+ uint32_t (*nrResidualSum)[MAX_NUM_TR_COEFFS];
+ uint32_t *nrCount;
};
class Quant
diff -r b015514a9386 -r b554b7cba38d source/encoder/encoder.cpp
--- a/source/encoder/encoder.cpp Sun Jul 26 16:12:32 2015 +0530
+++ b/source/encoder/encoder.cpp Mon Jul 27 17:14:23 2015 +0530
@@ -198,7 +198,6 @@
m_scalingList.setDefaultScalingList();
else if (m_scalingList.parseScalingList(m_param->scalingLists))
m_aborted = true;
- m_scalingList.setupQuantMatrices();
m_lookahead = new Lookahead(m_param, m_threadPool);
if (m_numPools)
@@ -214,6 +213,77 @@
initSPS(&m_sps);
initPPS(&m_pps);
+ uint16_t offsetEmergency[QP_MAX_MAX - QP_MAX_SPEC][MAX_NUM_TR_CATEGORIES][MAX_NUM_TR_COEFFS];
+
+ if (m_param->rc.vbvBufferSize)
+ {
+ bool scalingEnabled = m_scalingList.m_bEnabled;
+ if (!scalingEnabled)
+ {
+ m_scalingList.setDefaultScalingList();
+ m_scalingList.setupQuantMatrices();
+ }
+ else
+ m_scalingList.setupQuantMatrices();
+
+ for (int q = 0; q < QP_MAX_MAX - QP_MAX_SPEC; q++)
+ {
+ for (int cat = 0; cat < MAX_NUM_TR_CATEGORIES; cat++)
+ {
+ uint16_t *nrOffset = offsetEmergency[q][cat];
+
+ int trSize = cat & 3;
+
+ int coefCount = 1 << ((trSize + 2) * 2);
+
+ /* Denoise chroma first then luma, then DC. */
+ int dcThreshold = (QP_MAX_MAX - QP_MAX_SPEC) * 2 / 3;
+ int lumaThreshold = (QP_MAX_MAX - QP_MAX_SPEC) * 2 / 3;
+ int chromaThreshold = 0;
+
+ int thresh = (cat < 4 || (cat >= 8 && cat < 12)) ? lumaThreshold : chromaThreshold;
+
+ double quantF = (double)(1ULL << (q / 6 + 16 + 8));
+
+ for (int i = 0; i < coefCount; i++)
+ {
+ uint16_t max = (1 << (7 + X265_DEPTH)) - 1;
+ /* True "emergency mode": remove all DCT coefficients */
+ if (q == QP_MAX_MAX - QP_MAX_SPEC - 1)
+ {
+ nrOffset[i] = max;
+ continue;
+ }
+
+ int iThresh = i == 0 ? dcThreshold : thresh;
+ if (q < iThresh)
+ {
+ nrOffset[i] = 0;
+ continue;
+ }
+
+ int numList = (cat >= 8) * 3 + ((int)!iThresh);
+
+ double pos = (double)(q - iThresh + 1) / (QP_MAX_MAX - QP_MAX_SPEC - iThresh);
+ double start = quantF / (m_scalingList.m_quantCoef[trSize][numList][QP_MAX_SPEC % 6][i]);
+
+ // Formula chosen as an exponential scale to vaguely mimic the effects of a higher quantizer.
+ double bias = (pow(2, pos * (QP_MAX_MAX - QP_MAX_SPEC)) * 0.003 - 0.003) * start;
+ nrOffset[i] = (uint16_t)X265_MIN(bias + 0.5, max);
+ }
+ }
+ }
+
+ if (!scalingEnabled)
+ {
+ m_scalingList.m_bEnabled = false;
+ m_scalingList.m_bDataPresent = false;
+ m_scalingList.setupQuantMatrices();
+ }
+ }
+ else
+ m_scalingList.setupQuantMatrices();
+
int numRows = (m_param->sourceHeight + g_maxCUSize - 1) / g_maxCUSize;
int numCols = (m_param->sourceWidth + g_maxCUSize - 1) / g_maxCUSize;
for (int i = 0; i < m_param->frameNumThreads; i++)
@@ -223,6 +293,8 @@
x265_log(m_param, X265_LOG_ERROR, "Unable to initialize frame encoder, aborting\n");
m_aborted = true;
}
+ if (m_frameEncoder[i]->m_nr && m_param->rc.vbvBufferSize)
+ memcpy(m_frameEncoder[i]->m_nr->offsetEmergency, offsetEmergency, sizeof(offsetEmergency));
}
for (int i = 0; i < m_param->frameNumThreads; i++)
diff -r b015514a9386 -r b554b7cba38d source/encoder/frameencoder.cpp
--- a/source/encoder/frameencoder.cpp Sun Jul 26 16:12:32 2015 +0530
+++ b/source/encoder/frameencoder.cpp Mon Jul 27 17:14:23 2015 +0530
@@ -135,7 +135,7 @@
ok &= m_rce.picTimingSEI && m_rce.hrdTiming;
}
- if (m_param->noiseReductionIntra || m_param->noiseReductionInter)
+ if (m_param->noiseReductionIntra || m_param->noiseReductionInter || m_param->rc.vbvBufferSize)
m_nr = X265_MALLOC(NoiseReduction, 1);
if (m_nr)
memset(m_nr, 0, sizeof(NoiseReduction));
@@ -250,6 +250,9 @@
{
m_tld[i].analysis.initSearch(*m_param, m_top->m_scalingList);
m_tld[i].analysis.create(m_tld);
+ if (m_tld[i].analysis.m_quant.m_frameNr && m_param->rc.vbvBufferSize)
+ for (int tid = 0; tid < m_param->frameNumThreads; tid++)
+ memcpy(m_tld[i].analysis.m_quant.m_frameNr[tid].offsetEmergency, m_nr->offsetEmergency, sizeof(m_nr->offsetEmergency));
}
for (int i = 0; i < m_pool->m_numProviders; i++)
@@ -361,11 +364,45 @@
}
}
+ int numTLD;
+ if (m_pool)
+ numTLD = m_param->bEnableWavefront ? m_pool->m_numWorkers : m_pool->m_numWorkers + m_pool->m_numProviders;
+ else
+ numTLD = 1;
+
/* Get the QP for this frame from rate control. This call may block until
* frames ahead of it in encode order have called rateControlEnd() */
int qp = m_top->m_rateControl->rateControlStart(m_frame, &m_rce, m_top);
m_rce.newQp = qp;
+ if (m_nr)
+ {
+ if (qp > QP_MAX_SPEC && m_frame->m_param->rc.vbvBufferSize)
+ {
+ for (int i = 0; i < numTLD; i++)
+ {
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrOffset = m_tld[i].analysis.m_quant.m_frameNr[m_jpId].offsetEmergency[qp - QP_MAX_SPEC - 1];
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrResidualSum = m_tld[i].analysis.m_quant.m_frameNr[m_jpId].residualSumEmergency;
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrCount = m_tld[i].analysis.m_quant.m_frameNr[m_jpId].countEmergency;
+ }
+ }
+ else
+ {
+ if (m_param->noiseReductionIntra || m_param->noiseReductionInter)
+ {
+ for (int i = 0; i < numTLD; i++)
+ {
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrOffset = m_tld[i].analysis.m_quant.m_frameNr[m_jpId].offsetDenoise;
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrResidualSum = m_tld[i].analysis.m_quant.m_frameNr[m_jpId].residualSum;
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrCount = m_tld[i].analysis.m_quant.m_frameNr[m_jpId].count;
+ }
+ }
+ else
+ for (int i = 0; i < numTLD; i++)
+ m_tld[i].analysis.m_quant.m_frameNr[m_jpId].nrOffset = NULL;
+ }
+ }
+
/* Clip slice QP to 0-51 spec range before encoding */
slice->m_sliceQp = x265_clip3(-QP_BD_OFFSET, QP_MAX_SPEC, qp);
@@ -697,36 +734,35 @@
}
}
- int numTLD;
- if (m_pool)
- numTLD = m_param->bEnableWavefront ? m_pool->m_numWorkers : m_pool->m_numWorkers + m_pool->m_numProviders;
- else
- numTLD = 1;
-
if (m_nr)
{
- /* Accumulate NR statistics from all worker threads */
- for (int i = 0; i < numTLD; i++)
+ bool nrEnabled = (m_rce.newQp < QP_MAX_SPEC) && (m_param->noiseReductionIntra || m_param->noiseReductionInter);
+
+ if (nrEnabled)
{
- NoiseReduction* nr = &m_tld[i].analysis.m_quant.m_frameNr[m_jpId];
- for (int cat = 0; cat < MAX_NUM_TR_CATEGORIES; cat++)
+ /* Accumulate NR statistics from all worker threads */
+ for (int i = 0; i < numTLD; i++)
{
- for (int coeff = 0; coeff < MAX_NUM_TR_COEFFS; coeff++)
- m_nr->residualSum[cat][coeff] += nr->residualSum[cat][coeff];
-
- m_nr->count[cat] += nr->count[cat];
+ NoiseReduction* nr = &m_tld[i].analysis.m_quant.m_frameNr[m_jpId];
+ for (int cat = 0; cat < MAX_NUM_TR_CATEGORIES; cat++)
+ {
+ for (int coeff = 0; coeff < MAX_NUM_TR_COEFFS; coeff++)
+ m_nr->residualSum[cat][coeff] += nr->residualSum[cat][coeff];
+
+ m_nr->count[cat] += nr->count[cat];
+ }
}
- }
- noiseReductionUpdate();
+ noiseReductionUpdate();
- /* Copy updated NR coefficients back to all worker threads */
- for (int i = 0; i < numTLD; i++)
- {
- NoiseReduction* nr = &m_tld[i].analysis.m_quant.m_frameNr[m_jpId];
- memcpy(nr->offsetDenoise, m_nr->offsetDenoise, sizeof(uint16_t) * MAX_NUM_TR_CATEGORIES * MAX_NUM_TR_COEFFS);
- memset(nr->count, 0, sizeof(uint32_t) * MAX_NUM_TR_CATEGORIES);
- memset(nr->residualSum, 0, sizeof(uint32_t) * MAX_NUM_TR_CATEGORIES * MAX_NUM_TR_COEFFS);
+ /* Copy updated NR coefficients back to all worker threads */
+ for (int i = 0; i < numTLD; i++)
+ {
+ NoiseReduction* nr = &m_tld[i].analysis.m_quant.m_frameNr[m_jpId];
+ memcpy(nr->offsetDenoise, m_nr->offsetDenoise, sizeof(uint16_t)* MAX_NUM_TR_CATEGORIES * MAX_NUM_TR_COEFFS);
+ memset(nr->count, 0, sizeof(uint32_t)* MAX_NUM_TR_CATEGORIES);
+ memset(nr->residualSum, 0, sizeof(uint32_t)* MAX_NUM_TR_CATEGORIES * MAX_NUM_TR_COEFFS);
+ }
}
}
diff -r b015514a9386 -r b554b7cba38d source/encoder/search.cpp
--- a/source/encoder/search.cpp Sun Jul 26 16:12:32 2015 +0530
+++ b/source/encoder/search.cpp Mon Jul 27 17:14:23 2015 +0530
@@ -80,7 +80,7 @@
m_me.init(param.searchMethod, param.subpelRefine, param.internalCsp);
bool ok = m_quant.init(param.rdoqLevel, param.psyRdoq, scalingList, m_entropyCoder);
- if (m_param->noiseReductionIntra || m_param->noiseReductionInter)
+ if (m_param->noiseReductionIntra || m_param->noiseReductionInter || m_param->rc.vbvBufferSize)
ok &= m_quant.allocNoiseReduction(param);
ok &= Predict::allocBuffers(param.internalCsp); /* sets m_hChromaShift & m_vChromaShift */
More information about the x265-devel
mailing list