third_party_ffmpeg/libavcodec/snow.h
Derek Buitenhuis f3af379b5c Merge commit '2862b63783b5556f7f3fb2d097629bc6879f833a'
* commit '2862b63783b5556f7f3fb2d097629bc6879f833a':
  lavc: Move prediction_method to codec private options

Merged-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2016-02-03 16:49:19 +00:00

733 lines
22 KiB
C

/*
* Copyright (C) 2004 Michael Niedermayer <michaelni@gmx.at>
* Copyright (C) 2006 Robert Edele <yartrebo@earthlink.net>
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#ifndef AVCODEC_SNOW_H
#define AVCODEC_SNOW_H
#include "libavutil/motion_vector.h"
#include "hpeldsp.h"
#include "me_cmp.h"
#include "qpeldsp.h"
#include "snow_dwt.h"
#include "rangecoder.h"
#include "mathops.h"
#define FF_MPV_OFFSET(x) (offsetof(MpegEncContext, x) + offsetof(SnowContext, m))
#include "mpegvideo.h"
#include "h264qpel.h"
#define MID_STATE 128
#define MAX_PLANES 4
#define QSHIFT 5
#define QROOT (1<<QSHIFT)
#define LOSSLESS_QLOG -128
#define FRAC_BITS 4
#define MAX_REF_FRAMES 8
#define LOG2_OBMC_MAX 8
#define OBMC_MAX (1<<(LOG2_OBMC_MAX))
typedef struct BlockNode{
int16_t mx;
int16_t my;
uint8_t ref;
uint8_t color[3];
uint8_t type;
//#define TYPE_SPLIT 1
#define BLOCK_INTRA 1
#define BLOCK_OPT 2
//#define TYPE_NOCOLOR 4
uint8_t level; //FIXME merge into type?
}BlockNode;
static const BlockNode null_block= { //FIXME add border maybe
.color= {128,128,128},
.mx= 0,
.my= 0,
.ref= 0,
.type= 0,
.level= 0,
};
#define LOG2_MB_SIZE 4
#define MB_SIZE (1<<LOG2_MB_SIZE)
#define ENCODER_EXTRA_BITS 4
#define HTAPS_MAX 8
typedef struct x_and_coeff{
int16_t x;
uint16_t coeff;
} x_and_coeff;
typedef struct SubBand{
int level;
int stride;
int width;
int height;
int qlog; ///< log(qscale)/log[2^(1/6)]
DWTELEM *buf;
IDWTELEM *ibuf;
int buf_x_offset;
int buf_y_offset;
int stride_line; ///< Stride measured in lines, not pixels.
x_and_coeff * x_coeff;
struct SubBand *parent;
uint8_t state[/*7*2*/ 7 + 512][32];
}SubBand;
typedef struct Plane{
int width;
int height;
SubBand band[MAX_DECOMPOSITIONS][4];
int htaps;
int8_t hcoeff[HTAPS_MAX/2];
int diag_mc;
int fast_mc;
int last_htaps;
int8_t last_hcoeff[HTAPS_MAX/2];
int last_diag_mc;
}Plane;
typedef struct SnowContext{
AVClass *class;
AVCodecContext *avctx;
RangeCoder c;
MECmpContext mecc;
HpelDSPContext hdsp;
QpelDSPContext qdsp;
VideoDSPContext vdsp;
H264QpelContext h264qpel;
MpegvideoEncDSPContext mpvencdsp;
SnowDWTContext dwt;
AVFrame *input_picture; ///< new_picture with the internal linesizes
AVFrame *current_picture;
AVFrame *last_picture[MAX_REF_FRAMES];
uint8_t *halfpel_plane[MAX_REF_FRAMES][4][4];
AVFrame *mconly_picture;
// uint8_t q_context[16];
uint8_t header_state[32];
uint8_t block_state[128 + 32*128];
int keyframe;
int always_reset;
int version;
int spatial_decomposition_type;
int last_spatial_decomposition_type;
int temporal_decomposition_type;
int spatial_decomposition_count;
int last_spatial_decomposition_count;
int temporal_decomposition_count;
int max_ref_frames;
int ref_frames;
int16_t (*ref_mvs[MAX_REF_FRAMES])[2];
uint32_t *ref_scores[MAX_REF_FRAMES];
DWTELEM *spatial_dwt_buffer;
DWTELEM *temp_dwt_buffer;
IDWTELEM *spatial_idwt_buffer;
IDWTELEM *temp_idwt_buffer;
int *run_buffer;
int colorspace_type;
int chroma_h_shift;
int chroma_v_shift;
int spatial_scalability;
int qlog;
int last_qlog;
int lambda;
int lambda2;
int pass1_rc;
int mv_scale;
int last_mv_scale;
int qbias;
int last_qbias;
#define QBIAS_SHIFT 3
int b_width;
int b_height;
int block_max_depth;
int last_block_max_depth;
int nb_planes;
Plane plane[MAX_PLANES];
BlockNode *block;
#define ME_CACHE_SIZE 1024
unsigned me_cache[ME_CACHE_SIZE];
unsigned me_cache_generation;
slice_buffer sb;
int memc_only;
int no_bitstream;
int intra_penalty;
int motion_est;
int iterative_dia_size;
int scenechange_threshold;
MpegEncContext m; // needed for motion estimation, should not be used for anything else, the idea is to eventually make the motion estimation independent of MpegEncContext, so this will be removed then (FIXME/XXX)
uint8_t *scratchbuf;
uint8_t *emu_edge_buffer;
AVMotionVector *avmv;
int avmv_index;
uint64_t encoding_error[AV_NUM_DATA_POINTERS];
int pred;
}SnowContext;
/* Tables */
extern const uint8_t * const ff_obmc_tab[4];
extern uint8_t ff_qexp[QROOT];
extern int ff_scale_mv_ref[MAX_REF_FRAMES][MAX_REF_FRAMES];
/* C bits used by mmx/sse2/altivec */
static av_always_inline void snow_interleave_line_header(int * i, int width, IDWTELEM * low, IDWTELEM * high){
(*i) = (width) - 2;
if (width & 1){
low[(*i)+1] = low[((*i)+1)>>1];
(*i)--;
}
}
static av_always_inline void snow_interleave_line_footer(int * i, IDWTELEM * low, IDWTELEM * high){
for (; (*i)>=0; (*i)-=2){
low[(*i)+1] = high[(*i)>>1];
low[*i] = low[(*i)>>1];
}
}
static av_always_inline void snow_horizontal_compose_lift_lead_out(int i, IDWTELEM * dst, IDWTELEM * src, IDWTELEM * ref, int width, int w, int lift_high, int mul, int add, int shift){
for(; i<w; i++){
dst[i] = src[i] - ((mul * (ref[i] + ref[i + 1]) + add) >> shift);
}
if((width^lift_high)&1){
dst[w] = src[w] - ((mul * 2 * ref[w] + add) >> shift);
}
}
static av_always_inline void snow_horizontal_compose_liftS_lead_out(int i, IDWTELEM * dst, IDWTELEM * src, IDWTELEM * ref, int width, int w){
for(; i<w; i++){
dst[i] = src[i] + ((ref[i] + ref[(i+1)]+W_BO + 4 * src[i]) >> W_BS);
}
if(width&1){
dst[w] = src[w] + ((2 * ref[w] + W_BO + 4 * src[w]) >> W_BS);
}
}
/* common code */
int ff_snow_common_init(AVCodecContext *avctx);
int ff_snow_common_init_after_header(AVCodecContext *avctx);
void ff_snow_common_end(SnowContext *s);
void ff_snow_release_buffer(AVCodecContext *avctx);
void ff_snow_reset_contexts(SnowContext *s);
int ff_snow_alloc_blocks(SnowContext *s);
int ff_snow_frame_start(SnowContext *s);
void ff_snow_pred_block(SnowContext *s, uint8_t *dst, uint8_t *tmp, ptrdiff_t stride,
int sx, int sy, int b_w, int b_h, const BlockNode *block,
int plane_index, int w, int h);
int ff_snow_get_buffer(SnowContext *s, AVFrame *frame);
/* common inline functions */
//XXX doublecheck all of them should stay inlined
static inline void snow_set_blocks(SnowContext *s, int level, int x, int y, int l, int cb, int cr, int mx, int my, int ref, int type){
const int w= s->b_width << s->block_max_depth;
const int rem_depth= s->block_max_depth - level;
const int index= (x + y*w) << rem_depth;
const int block_w= 1<<rem_depth;
BlockNode block;
int i,j;
block.color[0]= l;
block.color[1]= cb;
block.color[2]= cr;
block.mx= mx;
block.my= my;
block.ref= ref;
block.type= type;
block.level= level;
for(j=0; j<block_w; j++){
for(i=0; i<block_w; i++){
s->block[index + i + j*w]= block;
}
}
}
static inline void pred_mv(SnowContext *s, int *mx, int *my, int ref,
const BlockNode *left, const BlockNode *top, const BlockNode *tr){
if(s->ref_frames == 1){
*mx = mid_pred(left->mx, top->mx, tr->mx);
*my = mid_pred(left->my, top->my, tr->my);
}else{
const int *scale = ff_scale_mv_ref[ref];
*mx = mid_pred((left->mx * scale[left->ref] + 128) >>8,
(top ->mx * scale[top ->ref] + 128) >>8,
(tr ->mx * scale[tr ->ref] + 128) >>8);
*my = mid_pred((left->my * scale[left->ref] + 128) >>8,
(top ->my * scale[top ->ref] + 128) >>8,
(tr ->my * scale[tr ->ref] + 128) >>8);
}
}
static av_always_inline int same_block(BlockNode *a, BlockNode *b){
if((a->type&BLOCK_INTRA) && (b->type&BLOCK_INTRA)){
return !((a->color[0] - b->color[0]) | (a->color[1] - b->color[1]) | (a->color[2] - b->color[2]));
}else{
return !((a->mx - b->mx) | (a->my - b->my) | (a->ref - b->ref) | ((a->type ^ b->type)&BLOCK_INTRA));
}
}
//FIXME name cleanup (b_w, block_w, b_width stuff)
//XXX should we really inline it?
static av_always_inline void add_yblock(SnowContext *s, int sliced, slice_buffer *sb, IDWTELEM *dst, uint8_t *dst8, const uint8_t *obmc, int src_x, int src_y, int b_w, int b_h, int w, int h, int dst_stride, int src_stride, int obmc_stride, int b_x, int b_y, int add, int offset_dst, int plane_index){
const int b_width = s->b_width << s->block_max_depth;
const int b_height= s->b_height << s->block_max_depth;
const int b_stride= b_width;
BlockNode *lt= &s->block[b_x + b_y*b_stride];
BlockNode *rt= lt+1;
BlockNode *lb= lt+b_stride;
BlockNode *rb= lb+1;
uint8_t *block[4];
// When src_stride is large enough, it is possible to interleave the blocks.
// Otherwise the blocks are written sequentially in the tmp buffer.
int tmp_step= src_stride >= 7*MB_SIZE ? MB_SIZE : MB_SIZE*src_stride;
uint8_t *tmp = s->scratchbuf;
uint8_t *ptmp;
int x,y;
if(b_x<0){
lt= rt;
lb= rb;
}else if(b_x + 1 >= b_width){
rt= lt;
rb= lb;
}
if(b_y<0){
lt= lb;
rt= rb;
}else if(b_y + 1 >= b_height){
lb= lt;
rb= rt;
}
if(src_x<0){ //FIXME merge with prev & always round internal width up to *16
obmc -= src_x;
b_w += src_x;
if(!sliced && !offset_dst)
dst -= src_x;
src_x=0;
}
if(src_x + b_w > w){
b_w = w - src_x;
}
if(src_y<0){
obmc -= src_y*obmc_stride;
b_h += src_y;
if(!sliced && !offset_dst)
dst -= src_y*dst_stride;
src_y=0;
}
if(src_y + b_h> h){
b_h = h - src_y;
}
if(b_w<=0 || b_h<=0) return;
if(!sliced && offset_dst)
dst += src_x + src_y*dst_stride;
dst8+= src_x + src_y*src_stride;
// src += src_x + src_y*src_stride;
ptmp= tmp + 3*tmp_step;
block[0]= ptmp;
ptmp+=tmp_step;
ff_snow_pred_block(s, block[0], tmp, src_stride, src_x, src_y, b_w, b_h, lt, plane_index, w, h);
if(same_block(lt, rt)){
block[1]= block[0];
}else{
block[1]= ptmp;
ptmp+=tmp_step;
ff_snow_pred_block(s, block[1], tmp, src_stride, src_x, src_y, b_w, b_h, rt, plane_index, w, h);
}
if(same_block(lt, lb)){
block[2]= block[0];
}else if(same_block(rt, lb)){
block[2]= block[1];
}else{
block[2]= ptmp;
ptmp+=tmp_step;
ff_snow_pred_block(s, block[2], tmp, src_stride, src_x, src_y, b_w, b_h, lb, plane_index, w, h);
}
if(same_block(lt, rb) ){
block[3]= block[0];
}else if(same_block(rt, rb)){
block[3]= block[1];
}else if(same_block(lb, rb)){
block[3]= block[2];
}else{
block[3]= ptmp;
ff_snow_pred_block(s, block[3], tmp, src_stride, src_x, src_y, b_w, b_h, rb, plane_index, w, h);
}
if(sliced){
s->dwt.inner_add_yblock(obmc, obmc_stride, block, b_w, b_h, src_x,src_y, src_stride, sb, add, dst8);
}else{
for(y=0; y<b_h; y++){
//FIXME ugly misuse of obmc_stride
const uint8_t *obmc1= obmc + y*obmc_stride;
const uint8_t *obmc2= obmc1+ (obmc_stride>>1);
const uint8_t *obmc3= obmc1+ obmc_stride*(obmc_stride>>1);
const uint8_t *obmc4= obmc3+ (obmc_stride>>1);
for(x=0; x<b_w; x++){
int v= obmc1[x] * block[3][x + y*src_stride]
+obmc2[x] * block[2][x + y*src_stride]
+obmc3[x] * block[1][x + y*src_stride]
+obmc4[x] * block[0][x + y*src_stride];
v <<= 8 - LOG2_OBMC_MAX;
if(FRAC_BITS != 8){
v >>= 8 - FRAC_BITS;
}
if(add){
v += dst[x + y*dst_stride];
v = (v + (1<<(FRAC_BITS-1))) >> FRAC_BITS;
if(v&(~255)) v= ~(v>>31);
dst8[x + y*src_stride] = v;
}else{
dst[x + y*dst_stride] -= v;
}
}
}
}
}
static av_always_inline void predict_slice(SnowContext *s, IDWTELEM *buf, int plane_index, int add, int mb_y){
Plane *p= &s->plane[plane_index];
const int mb_w= s->b_width << s->block_max_depth;
const int mb_h= s->b_height << s->block_max_depth;
int x, y, mb_x;
int block_size = MB_SIZE >> s->block_max_depth;
int block_w = plane_index ? block_size>>s->chroma_h_shift : block_size;
int block_h = plane_index ? block_size>>s->chroma_v_shift : block_size;
const uint8_t *obmc = plane_index ? ff_obmc_tab[s->block_max_depth+s->chroma_h_shift] : ff_obmc_tab[s->block_max_depth];
const int obmc_stride= plane_index ? (2*block_size)>>s->chroma_h_shift : 2*block_size;
int ref_stride= s->current_picture->linesize[plane_index];
uint8_t *dst8= s->current_picture->data[plane_index];
int w= p->width;
int h= p->height;
av_assert2(s->chroma_h_shift == s->chroma_v_shift); // obmc params assume squares
if(s->keyframe || (s->avctx->debug&512)){
if(mb_y==mb_h)
return;
if(add){
for(y=block_h*mb_y; y<FFMIN(h,block_h*(mb_y+1)); y++){
for(x=0; x<w; x++){
int v= buf[x + y*w] + (128<<FRAC_BITS) + (1<<(FRAC_BITS-1));
v >>= FRAC_BITS;
if(v&(~255)) v= ~(v>>31);
dst8[x + y*ref_stride]= v;
}
}
}else{
for(y=block_h*mb_y; y<FFMIN(h,block_h*(mb_y+1)); y++){
for(x=0; x<w; x++){
buf[x + y*w]-= 128<<FRAC_BITS;
}
}
}
return;
}
for(mb_x=0; mb_x<=mb_w; mb_x++){
add_yblock(s, 0, NULL, buf, dst8, obmc,
block_w*mb_x - block_w/2,
block_h*mb_y - block_h/2,
block_w, block_h,
w, h,
w, ref_stride, obmc_stride,
mb_x - 1, mb_y - 1,
add, 1, plane_index);
}
}
static av_always_inline void predict_plane(SnowContext *s, IDWTELEM *buf, int plane_index, int add){
const int mb_h= s->b_height << s->block_max_depth;
int mb_y;
for(mb_y=0; mb_y<=mb_h; mb_y++)
predict_slice(s, buf, plane_index, add, mb_y);
}
static inline void set_blocks(SnowContext *s, int level, int x, int y, int l, int cb, int cr, int mx, int my, int ref, int type){
const int w= s->b_width << s->block_max_depth;
const int rem_depth= s->block_max_depth - level;
const int index= (x + y*w) << rem_depth;
const int block_w= 1<<rem_depth;
const int block_h= 1<<rem_depth; //FIXME "w!=h"
BlockNode block;
int i,j;
block.color[0]= l;
block.color[1]= cb;
block.color[2]= cr;
block.mx= mx;
block.my= my;
block.ref= ref;
block.type= type;
block.level= level;
for(j=0; j<block_h; j++){
for(i=0; i<block_w; i++){
s->block[index + i + j*w]= block;
}
}
}
static inline void init_ref(MotionEstContext *c, uint8_t *src[3], uint8_t *ref[3], uint8_t *ref2[3], int x, int y, int ref_index){
SnowContext *s = c->avctx->priv_data;
const int offset[3]= {
y*c-> stride + x,
((y*c->uvstride + x)>>s->chroma_h_shift),
((y*c->uvstride + x)>>s->chroma_h_shift),
};
int i;
for(i=0; i<3; i++){
c->src[0][i]= src [i];
c->ref[0][i]= ref [i] + offset[i];
}
av_assert2(!ref_index);
}
/* bitstream functions */
extern const int8_t ff_quant3bA[256];
#define QEXPSHIFT (7-FRAC_BITS+8) //FIXME try to change this to 0
static inline void put_symbol(RangeCoder *c, uint8_t *state, int v, int is_signed){
int i;
if(v){
const int a= FFABS(v);
const int e= av_log2(a);
const int el= FFMIN(e, 10);
put_rac(c, state+0, 0);
for(i=0; i<el; i++){
put_rac(c, state+1+i, 1); //1..10
}
for(; i<e; i++){
put_rac(c, state+1+9, 1); //1..10
}
put_rac(c, state+1+FFMIN(i,9), 0);
for(i=e-1; i>=el; i--){
put_rac(c, state+22+9, (a>>i)&1); //22..31
}
for(; i>=0; i--){
put_rac(c, state+22+i, (a>>i)&1); //22..31
}
if(is_signed)
put_rac(c, state+11 + el, v < 0); //11..21
}else{
put_rac(c, state+0, 1);
}
}
static inline int get_symbol(RangeCoder *c, uint8_t *state, int is_signed){
if(get_rac(c, state+0))
return 0;
else{
int i, e, a;
e= 0;
while(get_rac(c, state+1 + FFMIN(e,9))){ //1..10
e++;
if (e > 31)
return AVERROR_INVALIDDATA;
}
a= 1;
for(i=e-1; i>=0; i--){
a += a + get_rac(c, state+22 + FFMIN(i,9)); //22..31
}
e= -(is_signed && get_rac(c, state+11 + FFMIN(e,10))); //11..21
return (a^e)-e;
}
}
static inline void put_symbol2(RangeCoder *c, uint8_t *state, int v, int log2){
int i;
int r= log2>=0 ? 1<<log2 : 1;
av_assert2(v>=0);
av_assert2(log2>=-4);
while(v >= r){
put_rac(c, state+4+log2, 1);
v -= r;
log2++;
if(log2>0) r+=r;
}
put_rac(c, state+4+log2, 0);
for(i=log2-1; i>=0; i--){
put_rac(c, state+31-i, (v>>i)&1);
}
}
static inline int get_symbol2(RangeCoder *c, uint8_t *state, int log2){
int i;
int r= log2>=0 ? 1<<log2 : 1;
int v=0;
av_assert2(log2>=-4);
while(log2<28 && get_rac(c, state+4+log2)){
v+= r;
log2++;
if(log2>0) r+=r;
}
for(i=log2-1; i>=0; i--){
v+= get_rac(c, state+31-i)<<i;
}
return v;
}
static inline void unpack_coeffs(SnowContext *s, SubBand *b, SubBand * parent, int orientation){
const int w= b->width;
const int h= b->height;
int x,y;
int run, runs;
x_and_coeff *xc= b->x_coeff;
x_and_coeff *prev_xc= NULL;
x_and_coeff *prev2_xc= xc;
x_and_coeff *parent_xc= parent ? parent->x_coeff : NULL;
x_and_coeff *prev_parent_xc= parent_xc;
runs= get_symbol2(&s->c, b->state[30], 0);
if(runs-- > 0) run= get_symbol2(&s->c, b->state[1], 3);
else run= INT_MAX;
for(y=0; y<h; y++){
int v=0;
int lt=0, t=0, rt=0;
if(y && prev_xc->x == 0){
rt= prev_xc->coeff;
}
for(x=0; x<w; x++){
int p=0;
const int l= v;
lt= t; t= rt;
if(y){
if(prev_xc->x <= x)
prev_xc++;
if(prev_xc->x == x + 1)
rt= prev_xc->coeff;
else
rt=0;
}
if(parent_xc){
if(x>>1 > parent_xc->x){
parent_xc++;
}
if(x>>1 == parent_xc->x){
p= parent_xc->coeff;
}
}
if(/*ll|*/l|lt|t|rt|p){
int context= av_log2(/*FFABS(ll) + */3*(l>>1) + (lt>>1) + (t&~1) + (rt>>1) + (p>>1));
v=get_rac(&s->c, &b->state[0][context]);
if(v){
v= 2*(get_symbol2(&s->c, b->state[context + 2], context-4) + 1);
v+=get_rac(&s->c, &b->state[0][16 + 1 + 3 + ff_quant3bA[l&0xFF] + 3*ff_quant3bA[t&0xFF]]);
if ((uint16_t)v != v) {
av_log(s->avctx, AV_LOG_ERROR, "Coefficient damaged\n");
v = 1;
}
xc->x=x;
(xc++)->coeff= v;
}
}else{
if(!run){
if(runs-- > 0) run= get_symbol2(&s->c, b->state[1], 3);
else run= INT_MAX;
v= 2*(get_symbol2(&s->c, b->state[0 + 2], 0-4) + 1);
v+=get_rac(&s->c, &b->state[0][16 + 1 + 3]);
if ((uint16_t)v != v) {
av_log(s->avctx, AV_LOG_ERROR, "Coefficient damaged\n");
v = 1;
}
xc->x=x;
(xc++)->coeff= v;
}else{
int max_run;
run--;
v=0;
av_assert2(run >= 0);
if(y) max_run= FFMIN(run, prev_xc->x - x - 2);
else max_run= FFMIN(run, w-x-1);
if(parent_xc)
max_run= FFMIN(max_run, 2*parent_xc->x - x - 1);
av_assert2(max_run >= 0 && max_run <= run);
x+= max_run;
run-= max_run;
}
}
}
(xc++)->x= w+1; //end marker
prev_xc= prev2_xc;
prev2_xc= xc;
if(parent_xc){
if(y&1){
while(parent_xc->x != parent->width+1)
parent_xc++;
parent_xc++;
prev_parent_xc= parent_xc;
}else{
parent_xc= prev_parent_xc;
}
}
}
(xc++)->x= w+1; //end marker
}
#endif /* AVCODEC_SNOW_H */