h264 loop filter optimizations

Originally committed as revision 2909 to svn://svn.ffmpeg.org/ffmpeg/trunk
This commit is contained in:
Michael Niedermayer 2004-03-20 00:18:52 +00:00
parent 980a82b70b
commit 3ebc7e04de
2 changed files with 220 additions and 162 deletions

View File

@ -1163,6 +1163,12 @@ static inline int clip(int a, int amin, int amax)
return a;
}
static inline int clip_uint8(int a)
{
if (a&(~255)) return (-a)>>31;
else return a;
}
/* math */
extern const uint8_t ff_sqrt_tab[128];

View File

@ -2971,6 +2971,10 @@ static int decode_slice_header(H264Context *h){
//FIXME CABAC stuff
s->qscale = h->pps.init_qp + get_se_golomb(&s->gb); //slice_qp_delta
if(s->qscale<0 || s->qscale>51){
av_log(s->avctx, AV_LOG_ERROR, "QP %d out of range\n", s->qscale);
return -1;
}
//FIXME qscale / qp ... stuff
if(h->slice_type == SP_TYPE){
get_bits1(&s->gb); /* sp_for_switch_flag */
@ -2979,19 +2983,15 @@ static int decode_slice_header(H264Context *h){
get_se_golomb(&s->gb); /* slice_qs_delta */
}
h->disable_deblocking_filter_idc = 0;
h->slice_alpha_c0_offset = 0;
h->slice_beta_offset = 0;
if( h->pps.deblocking_filter_parameters_present ) {
h->disable_deblocking_filter_idc= get_ue_golomb(&s->gb);
if( h->disable_deblocking_filter_idc != 1 ) {
h->slice_alpha_c0_offset = get_se_golomb(&s->gb) << 1;
h->slice_beta_offset = get_se_golomb(&s->gb) << 1;
} else {
h->slice_alpha_c0_offset = 0;
h->slice_beta_offset = 0;
}
} else {
h->disable_deblocking_filter_idc = 0;
h->slice_alpha_c0_offset = 0;
h->slice_beta_offset = 0;
}
#if 0 //FMO
@ -3624,69 +3624,84 @@ static void filter_mb_edgev( H264Context *h, uint8_t *pix, int stride, int bS[4]
continue;
}
/* 4px edge length */
for( d = 0; d < 4; d++ ) {
const uint8_t p0 = pix[-1];
const uint8_t p1 = pix[-2];
const uint8_t p2 = pix[-3];
if( bS[i] < 4 ) {
const int tc0 = tc0_table[index_a][bS[i] - 1];
/* 4px edge length */
for( d = 0; d < 4; d++ ) {
const int p0 = pix[-1];
const int p1 = pix[-2];
const int p2 = pix[-3];
const int q0 = pix[0];
const int q1 = pix[1];
const int q2 = pix[2];
const uint8_t q0 = pix[0];
const uint8_t q1 = pix[1];
const uint8_t q2 = pix[2];
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
int tc = tc0;
int i_delta;
if( abs( p0 - q0 ) >= alpha ||
abs( p1 - p0 ) >= beta ||
abs( q1 - q0 ) >= beta ) {
if( ABS( p2 - p0 ) < beta ) {
pix[-2] = p1 + clip( ( p2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( p1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
}
if( ABS( q2 - q0 ) < beta ) {
pix[1] = q1 + clip( ( q2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( q1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
}
i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
pix[-1] = clip_uint8( p0 + i_delta ); /* p0' */
pix[0] = clip_uint8( q0 - i_delta ); /* q0' */
}
pix += stride;
continue;
}
}else{
/* 4px edge length */
for( d = 0; d < 4; d++ ) {
const int p0 = pix[-1];
const int p1 = pix[-2];
const int p2 = pix[-3];
if( bS[i] < 4 ) {
const int tc0 = tc0_table[index_a][bS[i] - 1];
int tc = tc0;
int i_delta;
const int q0 = pix[0];
const int q1 = pix[1];
const int q2 = pix[2];
if( abs( p2 - p0 ) < beta ) {
pix[-2] = p1 + clip( ( p2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( p1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
if(ABS( p0 - q0 ) < (( alpha >> 2 ) + 2 )){
if( ABS( p2 - p0 ) < beta)
{
const int p3 = pix[-4];
/* p0', p1', p2' */
pix[-1] = ( p2 + 2*p1 + 2*p0 + 2*q0 + q1 + 4 ) >> 3;
pix[-2] = ( p2 + p1 + p0 + q0 + 2 ) >> 2;
pix[-3] = ( 2*p3 + 3*p2 + p1 + p0 + q0 + 4 ) >> 3;
} else {
/* p0' */
pix[-1] = ( 2*p1 + p0 + q1 + 2 ) >> 2;
}
if( ABS( q2 - q0 ) < beta)
{
const int q3 = pix[3];
/* q0', q1', q2' */
pix[0] = ( p1 + 2*p0 + 2*q0 + 2*q1 + q2 + 4 ) >> 3;
pix[1] = ( p0 + q0 + q1 + q2 + 2 ) >> 2;
pix[2] = ( 2*q3 + 3*q2 + q1 + q0 + p0 + 4 ) >> 3;
} else {
/* q0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2;
}
}else{
/* p0', q0' */
pix[-1] = ( 2*p1 + p0 + q1 + 2 ) >> 2;
pix[ 0] = ( 2*q1 + q0 + p1 + 2 ) >> 2;
}
}
if( abs( q2 - q0 ) < beta ) {
pix[1] = q1 + clip( ( q2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( q1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
}
i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
pix[-1] = clip( p0 + i_delta, 0, 255 ); /* p0' */
pix[0] = clip( q0 - i_delta, 0, 255 ); /* q0' */
pix += stride;
}
else
{
const int c = abs( p0 - q0 ) < (( alpha >> 2 ) + 2 );
if( abs( p2 - p0 ) < beta && c )
{
const uint8_t p3 = pix[-4];
/* p0', p1', p2' */
pix[-1] = ( p2 + 2*p1 + 2*p0 + 2*q0 + q1 + 4 ) >> 3;
pix[-2] = ( p2 + p1 + p0 + q0 + 2 ) >> 2;
pix[-3] = ( 2*p3 + 3*p2 + p1 + p0 + q0 + 4 ) >> 3;
} else {
/* p0' */
pix[-1] = ( 2*p1 + p0 + q1 + 2 ) >> 2;
}
if( abs( q2 - q0 ) < beta && c )
{
const uint8_t q3 = pix[3];
/* q0', q1', q2' */
pix[0] = ( p1 + 2*p0 + 2*q0 + 2*q1 + q2 + 4 ) >> 3;
pix[1] = ( p0 + q0 + q1 + q2 + 2 ) >> 2;
pix[2] = ( 2*q3 + 3*q2 + q1 + q0 + p0 + 4 ) >> 3;
} else {
/* q0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2;
}
}
pix += stride;
}
}
}
@ -3702,32 +3717,42 @@ static void filter_mb_edgecv( H264Context *h, uint8_t *pix, int stride, int bS[4
continue;
}
/* 2px edge length (because we use same bS than the one for luma) */
for( d = 0; d < 2; d++ )
{
const uint8_t p0 = pix[-1];
const uint8_t p1 = pix[-2];
const uint8_t q0 = pix[0];
const uint8_t q1 = pix[1];
if( bS[i] < 4 ) {
const int tc = tc0_table[index_a][bS[i] - 1] + 1;
/* 2px edge length (because we use same bS than the one for luma) */
for( d = 0; d < 2; d++ ){
const int p0 = pix[-1];
const int p1 = pix[-2];
const int q0 = pix[0];
const int q1 = pix[1];
if( abs( p0 - q0 ) >= alpha ||
abs( p1 - p0 ) >= beta ||
abs( q1 - q0 ) >= beta ) {
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
const int i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
pix[-1] = clip_uint8( p0 + i_delta ); /* p0' */
pix[0] = clip_uint8( q0 - i_delta ); /* q0' */
}
pix += stride;
continue;
}
}else{
/* 2px edge length (because we use same bS than the one for luma) */
for( d = 0; d < 2; d++ ){
const int p0 = pix[-1];
const int p1 = pix[-2];
const int q0 = pix[0];
const int q1 = pix[1];
if( bS[i] < 4 ) {
const int tc = tc0_table[index_a][bS[i] - 1] + 1;
const int i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
pix[-1] = clip( p0 + i_delta, 0, 255 ); /* p0' */
pix[0] = clip( q0 - i_delta, 0, 255 ); /* q0' */
} else {
pix[-1] = ( 2*p1 + p0 + q1 + 2 ) >> 2; /* p0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2; /* q0' */
pix[-1] = ( 2*p1 + p0 + q1 + 2 ) >> 2; /* p0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2; /* q0' */
}
pix += stride;
}
pix += stride;
}
}
}
@ -3745,66 +3770,83 @@ static void filter_mb_edgeh( H264Context *h, uint8_t *pix, int stride, int bS[4]
continue;
}
/* 4px edge length */
for( d = 0; d < 4; d++ ) {
const uint8_t p0 = pix[-1*pix_next];
const uint8_t p1 = pix[-2*pix_next];
const uint8_t p2 = pix[-3*pix_next];
const uint8_t q0 = pix[0];
const uint8_t q1 = pix[1*pix_next];
const uint8_t q2 = pix[2*pix_next];
if( bS[i] < 4 ) {
const int tc0 = tc0_table[index_a][bS[i] - 1];
/* 4px edge length */
for( d = 0; d < 4; d++ ) {
const int p0 = pix[-1*pix_next];
const int p1 = pix[-2*pix_next];
const int p2 = pix[-3*pix_next];
const int q0 = pix[0];
const int q1 = pix[1*pix_next];
const int q2 = pix[2*pix_next];
if( abs( p0 - q0 ) >= alpha ||
abs( p1 - p0 ) >= beta ||
abs( q1 - q0 ) >= beta ) {
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
int tc = tc0;
int i_delta;
if( ABS( p2 - p0 ) < beta ) {
pix[-2*pix_next] = p1 + clip( ( p2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( p1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
}
if( ABS( q2 - q0 ) < beta ) {
pix[pix_next] = q1 + clip( ( q2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( q1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
}
i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
pix[-pix_next] = clip_uint8( p0 + i_delta ); /* p0' */
pix[0] = clip_uint8( q0 - i_delta ); /* q0' */
}
pix++;
continue;
}
}else{
/* 4px edge length */
for( d = 0; d < 4; d++ ) {
const int p0 = pix[-1*pix_next];
const int p1 = pix[-2*pix_next];
const int p2 = pix[-3*pix_next];
const int q0 = pix[0];
const int q1 = pix[1*pix_next];
const int q2 = pix[2*pix_next];
if( bS[i] < 4 ) {
const int tc0 = tc0_table[index_a][bS[i] - 1];
int tc = tc0;
int i_delta;
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
if( abs( p2 - p0 ) < beta ) {
pix[-2*pix_next] = p1 + clip( ( p2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( p1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
const int p3 = pix[-4*pix_next];
const int q3 = pix[ 3*pix_next];
if(ABS( p0 - q0 ) < (( alpha >> 2 ) + 2 )){
if( ABS( p2 - p0 ) < beta) {
/* p0', p1', p2' */
pix[-1*pix_next] = ( p2 + 2*p1 + 2*p0 + 2*q0 + q1 + 4 ) >> 3;
pix[-2*pix_next] = ( p2 + p1 + p0 + q0 + 2 ) >> 2;
pix[-3*pix_next] = ( 2*p3 + 3*p2 + p1 + p0 + q0 + 4 ) >> 3;
} else {
/* p0' */
pix[-1*pix_next] = ( 2*p1 + p0 + q1 + 2 ) >> 2;
}
if( ABS( q2 - q0 ) < beta) {
/* q0', q1', q2' */
pix[0*pix_next] = ( p1 + 2*p0 + 2*q0 + 2*q1 + q2 + 4 ) >> 3;
pix[1*pix_next] = ( p0 + q0 + q1 + q2 + 2 ) >> 2;
pix[2*pix_next] = ( 2*q3 + 3*q2 + q1 + q0 + p0 + 4 ) >> 3;
} else {
/* q0' */
pix[0*pix_next] = ( 2*q1 + q0 + p1 + 2 ) >> 2;
}
}else{
/* p0', q0' */
pix[-1*pix_next] = ( 2*p1 + p0 + q1 + 2 ) >> 2;
pix[ 0*pix_next] = ( 2*q1 + q0 + p1 + 2 ) >> 2;
}
}
if( abs( q2 - q0 ) < beta ) {
pix[pix_next] = q1 + clip( ( q2 + ( ( p0 + q0 + 1 ) >> 1 ) - ( q1 << 1 ) ) >> 1, -tc0, tc0 );
tc++;
}
i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
pix[-pix_next] = clip( p0 + i_delta, 0, 255 ); /* p0' */
pix[0] = clip( q0 - i_delta, 0, 255 ); /* q0' */
pix++;
}
else
{
const uint8_t p3 = pix[-4*pix_next];
const uint8_t q3 = pix[ 3*pix_next];
const int c = abs( p0 - q0 ) < (( alpha >> 2 ) + 2 );
if( abs( p2 - p0 ) < beta && c ) {
/* p0', p1', p2' */
pix[-1*pix_next] = ( p2 + 2*p1 + 2*p0 + 2*q0 + q1 + 4 ) >> 3;
pix[-2*pix_next] = ( p2 + p1 + p0 + q0 + 2 ) >> 2;
pix[-3*pix_next] = ( 2*p3 + 3*p2 + p1 + p0 + q0 + 4 ) >> 3;
} else {
/* p0' */
pix[-1*pix_next] = ( 2*p1 + p0 + q1 + 2 ) >> 2;
}
if( abs( q2 - q0 ) < beta && c ) {
/* q0', q1', q2' */
pix[0*pix_next] = ( p1 + 2*p0 + 2*q0 + 2*q1 + q2 + 4 ) >> 3;
pix[1*pix_next] = ( p0 + q0 + q1 + q2 + 2 ) >> 2;
pix[2*pix_next] = ( 2*q3 + 3*q2 + q1 + q0 + p0 + 4 ) >> 3;
} else {
/* q0' */
pix[0*pix_next] = ( 2*q1 + q0 + p1 + 2 ) >> 2;
}
}
pix++;
}
}
}
@ -3823,33 +3865,43 @@ static void filter_mb_edgech( H264Context *h, uint8_t *pix, int stride, int bS[4
continue;
}
/* 2px edge length (see deblocking_filter_edgecv) */
for( d = 0; d < 2; d++ ) {
const uint8_t p0 = pix[-1*pix_next];
const uint8_t p1 = pix[-2*pix_next];
const uint8_t q0 = pix[0];
const uint8_t q1 = pix[1*pix_next];
if( bS[i] < 4 ) {
int tc = tc0_table[index_a][bS[i] - 1] + 1;
/* 2px edge length (see deblocking_filter_edgecv) */
for( d = 0; d < 2; d++ ) {
const int p0 = pix[-1*pix_next];
const int p1 = pix[-2*pix_next];
const int q0 = pix[0];
const int q1 = pix[1*pix_next];
if( abs( p0 - q0 ) >= alpha ||
abs( p1 - p0 ) >= beta ||
abs( q1 - q0 ) >= beta ) {
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
int i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
pix[-pix_next] = clip_uint8( p0 + i_delta ); /* p0' */
pix[0] = clip_uint8( q0 - i_delta ); /* q0' */
}
pix++;
continue;
}
}else{
/* 2px edge length (see deblocking_filter_edgecv) */
for( d = 0; d < 2; d++ ) {
const int p0 = pix[-1*pix_next];
const int p1 = pix[-2*pix_next];
const int q0 = pix[0];
const int q1 = pix[1*pix_next];
if( bS[i] < 4 ) {
int tc = tc0_table[index_a][bS[i] - 1] + 1;
int i_delta = clip( (((q0 - p0 ) << 2) + (p1 - q1) + 4) >> 3, -tc, tc );
if( ABS( p0 - q0 ) < alpha &&
ABS( p1 - p0 ) < beta &&
ABS( q1 - q0 ) < beta ) {
pix[-pix_next] = clip( p0 + i_delta, 0, 255 ); /* p0' */
pix[0] = clip( q0 - i_delta, 0, 255 ); /* q0' */
pix[-pix_next] = ( 2*p1 + p0 + q1 + 2 ) >> 2; /* p0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2; /* q0' */
}
pix++;
}
else
{
pix[-pix_next] = ( 2*p1 + p0 + q1 + 2 ) >> 2; /* p0' */
pix[0] = ( 2*q1 + q0 + p1 + 2 ) >> 2; /* q0' */
}
pix++;
}
}
}
@ -3917,13 +3969,13 @@ static void filter_mb( H264Context *h, int mb_x, int mb_y ) {
bS[i] = 2;
}
else if( h->slice_type == P_TYPE ) {
const int b8_xy = h->mb2b8_xy[mb_xy]+(y/2)*h->b8_stride+(x/2);
const int b8n_xy= h->mb2b8_xy[mbn_xy]+(yn/2)*h->b8_stride+(xn/2);
const int b8_xy = h->mb2b8_xy[mb_xy]+(y>>1)*h->b8_stride+(x>>1);
const int b8n_xy= h->mb2b8_xy[mbn_xy]+(yn>>1)*h->b8_stride+(xn>>1);
const int b_xy = h->mb2b_xy[mb_xy]+y*h->b_stride+x;
const int bn_xy = h->mb2b_xy[mbn_xy]+yn*h->b_stride+xn;
if( s->current_picture.ref_index[0][b8_xy] != s->current_picture.ref_index[0][b8n_xy] ||
abs( s->current_picture.motion_val[0][b_xy][0] - s->current_picture.motion_val[0][bn_xy][0] ) >= 4 ||
abs( s->current_picture.motion_val[0][b_xy][1] - s->current_picture.motion_val[0][bn_xy][1] ) >= 4 )
ABS( s->current_picture.motion_val[0][b_xy][0] - s->current_picture.motion_val[0][bn_xy][0] ) >= 4 ||
ABS( s->current_picture.motion_val[0][b_xy][1] - s->current_picture.motion_val[0][bn_xy][1] ) >= 4 )
bS[i] = 1;
else
bS[i] = 0;
@ -3939,7 +3991,7 @@ static void filter_mb( H264Context *h, int mb_x, int mb_y ) {
qp = ( s->current_picture.qscale_table[mb_xy] + s->current_picture.qscale_table[mbn_xy] + 1 ) >> 1;
if( dir == 0 ) {
filter_mb_edgev( h, &img_y[4*edge], linesize, bS, qp );
if( edge%2 == 0 ) {
if( (edge&1) == 0 ) {
int chroma_qp = ( get_chroma_qp( h, s->current_picture.qscale_table[mb_xy] ) +
get_chroma_qp( h, s->current_picture.qscale_table[mbn_xy] ) + 1 ) >> 1;
filter_mb_edgecv( h, &img_cb[2*edge], uvlinesize, bS, chroma_qp );
@ -3947,7 +3999,7 @@ static void filter_mb( H264Context *h, int mb_x, int mb_y ) {
}
} else {
filter_mb_edgeh( h, &img_y[4*edge*linesize], linesize, bS, qp );
if( edge%2 == 0 ) {
if( (edge&1) == 0 ) {
int chroma_qp = ( get_chroma_qp( h, s->current_picture.qscale_table[mb_xy] ) +
get_chroma_qp( h, s->current_picture.qscale_table[mbn_xy] ) + 1 ) >> 1;
filter_mb_edgech( h, &img_cb[2*edge*uvlinesize], uvlinesize, bS, chroma_qp );