#include <assert.h>
#include <math.h>
#include "../rfxswf.h"
-#include "h263tables.c"
+#include "h263tables.h"
+#include "dct.h"
+
+/* TODO:
+ - use prepare* / write* in encode_IFrame_block
+ - check whether mvd steps of 2 lead to (much) smaller results
+*/
#ifdef MAIN
U16 totalframes = 0;
swf_SetU8(tag, 0); /* smoothing off */
swf_SetU8(tag, 2); /* codec = h.263 sorenson spark */
+#ifdef MAIN
totalframes = frames;
-
+#endif
memset(stream, 0, sizeof(VIDEOSTREAM));
stream->olinex = width;
+ stream->owidth = width;
+ stream->oheight = height;
width+=15;width&=~15;
height+=15;height&=~15;
stream->linex = width;
memset(stream->oldpic, 0, width*height*sizeof(YUV));
memset(stream->current, 0, width*height*sizeof(YUV));
+
+ assert((stream->width&15) == 0);
+ assert((stream->height&15) == 0);
+ assert((stream->bbx*16) == stream->width);
+ assert((stream->bby*16) == stream->height);
}
void swf_VideoStreamClear(VIDEOSTREAM*stream)
{
int v[64];
} block_t;
-static int zigzagtable[64] = {
- 0, 1, 5, 6, 14, 15, 27, 28,
- 2, 4, 7, 13, 16, 26, 29, 42,
- 3, 8, 12, 17, 25, 30, 41, 43,
- 9, 11, 18, 24, 31, 40, 44, 53,
- 10, 19, 23, 32, 39, 45, 52, 54,
- 20, 22, 33, 38, 46, 51, 55, 60,
- 21, 34, 37, 47, 50, 56, 59, 61,
- 35, 36, 48, 49, 57, 58, 62, 63};
-
-static void zigzag(int*src)
-{
- int tmp[64];
- int t;
- for(t=0;t<64;t++) {
- tmp[zigzagtable[t]] = src[t];
- }
- memcpy(src, tmp, sizeof(int)*64);
-}
-
-#define PI 3.14159265358979
-#define SQRT2 1.414214
-#define RSQRT2 (1.0/1.414214)
-
-static double table[8][8] =
-{
-{0.707106781186548,0.707106781186548,0.707106781186548,0.707106781186548,0.707106781186548,0.707106781186548,0.707106781186548,0.707106781186548},
-{0.980785280403230,0.831469612302545,0.555570233019602,0.195090322016128,-0.195090322016128,-0.555570233019602,-0.831469612302545,-0.980785280403230},
-{0.923879532511287,0.382683432365090,-0.382683432365090,-0.923879532511287,-0.923879532511287,-0.382683432365090,0.382683432365090,0.923879532511287},
-{0.831469612302545,-0.195090322016128,-0.980785280403230,-0.555570233019602,0.555570233019602,0.980785280403230,0.195090322016129,-0.831469612302545},
-{0.707106781186548,-0.707106781186547,-0.707106781186548,0.707106781186547,0.707106781186548,-0.707106781186547,-0.707106781186547,0.707106781186547},
-{0.555570233019602,-0.980785280403230,0.195090322016128,0.831469612302545,-0.831469612302545,-0.195090322016128,0.980785280403231,-0.555570233019602},
-{0.382683432365090,-0.923879532511287,0.923879532511287,-0.382683432365090,-0.382683432365091,0.923879532511287,-0.923879532511286,0.382683432365090},
-{0.195090322016128,-0.555570233019602,0.831469612302545,-0.980785280403231,0.980785280403230,-0.831469612302545,0.555570233019602,-0.195090322016129}
-};
-
-static void dct(int*src)
-{
- double tmp[64];
- int x,y,u,v,t;
-
- for(v=0;v<8;v++)
- for(u=0;u<8;u++)
- {
- double c = 0;
- for(x=0;x<8;x++)
- {
- c+=table[u][x]*src[v*8+x];
- }
- tmp[v*8+u] = c;
- }
- for(u=0;u<8;u++)
- for(v=0;v<8;v++)
- {
- double c = 0;
- for(y=0;y<8;y++)
- {
- c+=table[v][y]*tmp[y*8+u];
- }
- src[v*8+u] = (int)(c*0.25+0.5);
- }
-}
-
-static void idct(int*src)
-{
- double tmp[64];
- int x,y,u,v;
- for(y=0;y<8;y++)
- for(x=0;x<8;x++)
- {
- double c = 0;
- for(u=0;u<8;u++)
- {
- c+=table[u][x]*src[y*8+u];
- }
- tmp[y*8+x] = c;
- }
- for(y=0;y<8;y++)
- for(x=0;x<8;x++)
- {
- double c = 0;
- for(v=0;v<8;v++)
- {
- c+=table[v][y]*tmp[v*8+x];
- }
- src[y*8+x] = (int)(c*0.25+0.5);
- }
-}
-
-static double c[8] = {1.0,
-0.980785280403230, // cos(Pi*1/16), sin(Pi*7/16)
-0.923879532511287, // cos(Pi*2/16), sin(Pi*6/16)
-0.831469612302545, // cos(Pi*3/16), sin(Pi*5/16)
-0.707106781186548, // cos(Pi*4/16), sin(Pi*4/16), 1/sqrt(2)
-0.555570233019602, // cos(Pi*5/16), sin(Pi*3/16)
-0.382683432365090, // cos(Pi*6/16), sin(Pi*2/16)
-0.195090322016128 // cos(Pi*7/16), sin(Pi*1/16)
-};
-
-static double cc[8];
-int ccquant = -1;
-
-static void preparequant(int quant)
-{
- if(ccquant == quant)
- return;
- cc[0] = c[0]/(quant*2*4);
- cc[1] = c[1]/(quant*2*4);
- cc[2] = c[2]/(quant*2*4);
- cc[3] = c[3]/(quant*2*4);
- cc[4] = c[4]/(quant*2*4);
- cc[5] = c[5]/(quant*2*4);
- cc[6] = c[6]/(quant*2*4);
- cc[7] = c[7]/(quant*2*4);
- ccquant = quant;
-}
-
-inline static void innerdct(double*a,double*b, double*c)
-{
- // c1*c7*2 = c6
- // c2*c6*2 = c4
- // c3*c5*2 = c2
- // c4*c4*2 = 1
-
- //{ 1, 3, 5, 7, -7, -5, -3, -1},
- //{ 3, -7, -1, -5, 5, 1, 7, -3},
- //{ 5, -1, 7, 3, -3, -7, 1, -5},
- //{ 7, -5, 3, -1, 1, -3, 5, -7}
- double b0,b1,b2,b3,b4,b5;
- b2 = (a[0]+a[7]);
- b3 = (a[1]+a[6]);
- b4 = (a[2]+a[5]);
- b5 = (a[3]+a[4]);
-
- b0 = (b2+b5)*c[4];
- b1 = (b3+b4)*c[4];
- b[0*8] = b0 + b1;
- b[4*8] = b0 - b1;
- b[2*8] = (b2-b5)*c[2] + (b3-b4)*c[6];
- b[6*8] = (b2-b5)*c[6] + (b4-b3)*c[2];
-
- b0 = (a[0]-a[7]);
- b1 = (a[1]-a[6]);
- b2 = (a[2]-a[5]);
- b3 = (a[3]-a[4]);
-
- b[1*8] = b0*c[1] + b1*c[3] + b2*c[5] + b3*c[7];
- b[3*8] = b0*c[3] - b1*c[7] - b2*c[1] - b3*c[5];
- b[5*8] = b0*c[5] - b1*c[1] + b2*c[7] + b3*c[3];
- b[7*8] = b0*c[7] - b1*c[5] + b2*c[3] - b3*c[1];
-}
-
-static void dct2(int*src, int*dest)
-{
- double tmp[64], tmp2[64];
- double*p;
- int u,x,v,t;
-
- for(t=0;t<64;t++)
- tmp2[t] = src[t];
-
- for(v=0;v<8;v++)
- {
- double* a=&tmp2[v*8];
- double* b=&tmp[v];
- innerdct(a,b,c);
- }
- for(v=0;v<8;v++)
- {
- double* a=&tmp[v*8];
- double* b=&tmp2[v];
- innerdct(a,b,cc);
- }
- for(t=0;t<64;t++) {
- dest[zigzagtable[t]] = (int)(tmp2[t]);
- }
-}
-
-
static inline int truncate256(int a)
{
if(a>255) return 255;
{
YUV*p1;
YUV*p2;
- int y1=0, y2=0, y3=0, y4=0;
- int u=0,v=0;
+ int i=0;
int x,y;
posx*=16;
posy*=16;
p2 = p1;
for(y=0;y<8;y++) {
for(x=0;x<8;x++) {
- bb->u[u++] = (p2[x*2].u + p2[x*2+1].u + p2[linex+x*2].u + p2[linex+x*2+1].u)/4;
- bb->v[v++] = (p2[x*2].v + p2[x*2+1].v + p2[linex+x*2].v + p2[linex+x*2+1].v)/4;
- bb->y1[y1++] = p1[x].y;
- bb->y2[y2++] = p1[x+8].y;
- bb->y3[y3++] = p1[linex*8+x].y;
- bb->y4[y4++] = p1[linex*8+x+8].y;
+ bb->u[i] = (p2[x*2].u + p2[x*2+1].u + p2[linex+x*2].u + p2[linex+x*2+1].u)/4;
+ bb->v[i] = (p2[x*2].v + p2[x*2+1].v + p2[linex+x*2].v + p2[linex+x*2+1].v)/4;
+ bb->y1[i] = p1[x].y;
+ bb->y2[i] = p1[x+8].y;
+ bb->y3[i] = p1[linex*8+x].y;
+ bb->y4[i] = p1[linex*8+x+8].y;
+ i++;
}
p1+=linex;
p2+=linex*2;
}
}
+/* This function is pretty complex. Let's hope it works correctly */
static void getmvdregion(block_t* bb, YUV*pic, int posx, int posy, int mvdx, int mvdy, int linex)
{
YUV*p1;
int x,y;
int yhp = 0, uvhp=0;
int uvposx, uvposy;
- posx = posx*16 + ((mvdx&~1)/2);
+ posx = posx*16 + ((mvdx&~1)/2); //works also for negative mvdx (unlike mvdx/2)
posy = posy*16 + ((mvdy&~1)/2);
p1 = &pic[posy*linex+posx];
p2 = &pic[(posy&~1)*linex+(posx&~1)];
}
}
}
+
static void copyregion(VIDEOSTREAM*s, YUV*dest, YUV*src, int bx, int by)
{
- YUV*p1 = &src[by*s->linex*16+bx*16];
- YUV*p2 = &dest[by*s->linex*16+bx*16];
+ YUV*p1 = &dest[by*s->linex*16+bx*16];
+ YUV*p2 = &src[by*s->linex*16+bx*16];
int y;
for(y=0;y<16;y++) {
memcpy(p1, p2, 16*sizeof(YUV));
}
}
}
-static void copyblock(VIDEOSTREAM*s, YUV*dest, block_t*b, int bx, int by)
+static void copy_block_pic(VIDEOSTREAM*s, YUV*dest, block_t*b, int bx, int by)
{
YUV*p1 = &dest[(by*16)*s->linex+bx*16];
YUV*p2 = &dest[(by*16+8)*s->linex+bx*16];
}
}
-static int compareregions(VIDEOSTREAM*s, int bx, int by)
+static int compare_pic_pic(VIDEOSTREAM*s, YUV*pp1, YUV*pp2, int bx, int by)
{
int linex = s->width;
- YUV*p1 = &s->current[by*linex*16+bx*16];
- YUV*p2 = &s->oldpic[by*linex*16+bx*16];
- int diff = 0;
+ YUV*p1 = &pp1[by*linex*16+bx*16];
+ YUV*p2 = &pp2[by*linex*16+bx*16];
+ int diffy=0, diffuv = 0;
int x,y;
for(y=0;y<16;y++) {
for(x=0;x<16;x++) {
int y = m->y - n->y;
int u = m->u - n->u;
int v = m->v - n->v;
- diff += y*y+(u*u+v*v)/4;
+ diffy += abs(y);
+ diffuv += abs(u)+abs(v);
}
p1+=linex;
p2+=linex;
}
- return diff/256;
+ return diffy + diffuv/4;
+}
+
+static int compare_pic_block(VIDEOSTREAM*s, block_t* b, YUV*pic, int bx, int by)
+{
+ int linex = s->width;
+ YUV*y1 = &pic[(by*2)*linex*8+bx*16];
+ YUV*y2 = &pic[(by*2)*linex*8+bx*16+8];
+ YUV*y3 = &pic[(by*2+1)*linex*8+bx*16];
+ YUV*y4 = &pic[(by*2+1)*linex*8+bx*16+8];
+ YUV*uv1 = y1;
+ YUV*uv2 = &y1[linex];
+ int diffy=0, diffuv = 0;
+ int x,y;
+ for(y=0;y<8;y++) {
+ for(x=0;x<8;x++) {
+ int yy,u1,v1,u2,v2,u3,v3,u4,v4;
+ int y8x = y*8+x;
+ yy = y1[x].y - b->y1[y8x];
+ diffy += abs(yy);
+ yy = y2[x].y - b->y2[y8x];
+ diffy += abs(yy);
+ yy = y3[x].y - b->y3[y8x];
+ diffy += abs(yy);
+ yy = y4[x].y - b->y4[y8x];
+ diffy += abs(yy);
+ u1 = uv1[x*2].u - b->u[y8x];
+ v1 = uv1[x*2].v - b->v[y8x];
+ u2 = uv1[x*2+1].u - b->u[y8x];
+ v2 = uv1[x*2+1].v - b->v[y8x];
+ u3 = uv2[x*2].u - b->u[y8x];
+ v3 = uv2[x*2].v - b->v[y8x];
+ u4 = uv2[x*2+1].u - b->u[y8x];
+ v4 = uv2[x*2+1].v - b->v[y8x];
+ diffuv += (abs(u1)+abs(v1));
+ diffuv += (abs(u2)+abs(v2));
+ diffuv += (abs(u3)+abs(v3));
+ diffuv += (abs(u4)+abs(v4));
+ }
+ y1+=linex;
+ y2+=linex;
+ y3+=linex;
+ y4+=linex;
+ uv1+=linex*2;
+ uv2+=linex*2;
+ }
+ return diffy + diffuv/4;
}
static inline int valtodc(int val)
return val;
}
+/* TODO: we could also just let the caller pass only the string table[index] here */
static int codehuffman(TAG*tag, struct huffcode*table, int index)
{
/* TODO: !optimize! */
return bits;
}
-static void encode8x8(TAG*tag, int*bb, int has_dc, int has_tcoef)
+static int encode8x8(TAG*tag, int*bb, int has_dc, int has_tcoef)
{
int t;
int pos=0;
if(has_dc) {
swf_SetBits(tag, bb[0], 8);
+ bits += 8;
pos++;
}
if(rle_params[t].run == run &&
rle_params[t].level == level &&
rle_params[t].last == islast) {
- codehuffman(tag, rle, t);
+ bits += codehuffman(tag, rle, t);
swf_SetBits(tag, sign, 1);
+ bits += 1;
break;
}
}
if(t==RLE_ESCAPE) {
- codehuffman(tag, rle, RLE_ESCAPE);
+ bits += codehuffman(tag, rle, RLE_ESCAPE);
level=bb[pos];
/* table 14/h.263 */
assert(level);
swf_SetBits(tag, islast, 1);
swf_SetBits(tag, run, 6);
swf_SetBits(tag, level, 8); //FIXME: fixme??
+ bits += 1 + 6 + 8;
}
if(islast)
pos++;
}
}
+ return bits;
}
static void quantize(block_t*fb, block_t*b, int has_dc, int quant)
zigzag(fb->u);
zigzag(fb->v);
}
+
static void dodctandquant(block_t*fb, block_t*b, int has_dc, int quant)
{
int t;
preparequant(quant);
dct2(fb->y1,b->y1); dct2(fb->y2,b->y2); dct2(fb->y3,b->y3); dct2(fb->y4,b->y4);
dct2(fb->u,b->u); dct2(fb->v,b->v);
+
+ for(t=0;t<64;t++) {
+ /* prepare for encoding (only values in (-127..-1,1..127) are
+ allowed as non-zero, non-dc values */
+ if(b->y1[t]<-127) b->y1[t]=-127;
+ if(b->y2[t]<-127) b->y2[t]=-127;
+ if(b->y3[t]<-127) b->y3[t]=-127;
+ if(b->y4[t]<-127) b->y4[t]=-127;
+ if(b->u[t]<-127) b->u[t]=-127;
+ if(b->v[t]<-127) b->v[t]=-127;
+
+ if(b->y1[t]>127) b->y1[t]=127;
+ if(b->y2[t]>127) b->y2[t]=127;
+ if(b->y3[t]>127) b->y3[t]=127;
+ if(b->y4[t]>127) b->y4[t]=127;
+ if(b->u[t]>127) b->u[t]=127;
+ if(b->v[t]>127) b->v[t]=127;
+ }
}
static void doidct(block_t*b)
}
idct(fb.y1); idct(fb.y2); idct(fb.y3); idct(fb.y4);
idct(fb.u); idct(fb.v);
- for(t=0;t<64;t++) {
- b->y1[t] = fb.y1[t];
- b->y2[t] = fb.y2[t];
- b->y3[t] = fb.y3[t];
- b->y4[t] = fb.y4[t];
- b->u[t] = fb.u[t];
- b->v[t] = fb.v[t];
- }
+
+ memcpy(b, &fb, sizeof(block_t));
}
+
static void truncateblock(block_t*b)
{
int t;
*dquant = 0;
}
-static void encode_blockI(TAG*tag, VIDEOSTREAM*s, int bx, int by, int*quant)
-{
- block_t fb;
- block_t b;
- int dquant=0;
- int cbpcbits = 0, cbpybits=0;
-
- getregion(&fb, s->current, bx, by, s->width);
-
- change_quant(*quant, &dquant);
- *quant+=dquant;
-
- dodctandquant(&fb, &b, 1, *quant);
- //quantize(&fb, &b, 1, *quant);
-
- //decode_blockI(s, &b, bx, by);
-
- getblockpatterns(&b, &cbpybits, &cbpcbits, 1);
-
- if(dquant) {
- codehuffman(tag, mcbpc_intra, 4+cbpcbits);
- } else {
- codehuffman(tag, mcbpc_intra, 0+cbpcbits);
- }
-
- codehuffman(tag, cbpy, cbpybits);
-
- if(dquant) {
- setQuant(tag, dquant);
- }
-
- /* luminance */
- encode8x8(tag, b.y1, 1, cbpybits&8);
- encode8x8(tag, b.y2, 1, cbpybits&4);
- encode8x8(tag, b.y3, 1, cbpybits&2);
- encode8x8(tag, b.y4, 1, cbpybits&1);
-
- /* chrominance */
- encode8x8(tag, b.u, 1, cbpcbits&2);
- encode8x8(tag, b.v, 1, cbpcbits&1);
-
- /* reconstruct */
- dequantize(&b, 1, *quant);
- doidct(&b);
- truncateblock(&b);
- copyblock(s, s->current, &b, bx, by);
-}
-
static void yuvdiff(block_t*a, block_t*b)
{
int t;
} else if((x1 <= x3 && x3 <= x2) ||
(x2 <= x3 && x3 <= x1)) {
x4=x3;
+ } else {
+ x4=0;
+ assert(x4);
}
if((y1 <= y2 && y2 <= y3) ||
} else if((y1 <= y3 && y3 <= y2) ||
(y2 <= y3 && y3 <= y1)) {
y4=y3;
+ } else {
+ y4=0;
+ assert(y4);
}
*px = x4;
return x;
}
-static int encode_blockP(TAG*tag, VIDEOSTREAM*s, int bx, int by, int*quant)
+typedef struct _iblockdata_t
+{
+ block_t b; //transformed quantized coefficients
+ block_t reconstruction;
+ int bits;
+ int bx,by;
+ struct huffcode*ctable; //table to use for chrominance encoding (different for i-frames)
+ int iframe; // 1 if this is part of an iframe
+} iblockdata_t;
+
+typedef struct _mvdblockdata_t
{
- block_t fb;
block_t b;
- int dquant=0;
- int has_mvd=0;
- int has_mvd24=0;
- int has_dc=1;
- int mode = 0;
- int cbpcbits = 0, cbpybits=0;
- int diff;
- int predictmvdx;
- int predictmvdy;
-
- block_t b_i;
- int bits_i;
+ block_t fbold;
+ block_t reconstruction;
+ int xindex;
+ int yindex;
+ int movex;
+ int movey;
+ int bits;
+ int bx,by;
+} mvdblockdata_t;
+
+void prepareIBlock(VIDEOSTREAM*s, iblockdata_t*data, int bx, int by, block_t* fb, int*bits, int iframe)
+{
+ /* consider I-block */
+ block_t fb_i;
+ block_t b;
+ int y,c;
+ struct huffcode*ctable;
- block_t fbold_v00;
- block_t b_v00;
- int bits_v00 = 65535;
- int x_v00=0;
- int y_v00=0;
+ data->bx = bx;
+ data->by = by;
- diff = compareregions(s, bx, by);
+ data->iframe = iframe;
+ if(!iframe) {
+ data->ctable = &mcbpc_inter[3*4];
+ } else {
+ data->ctable = &mcbpc_intra[0];
+ }
- if(diff < *quant) {
- /* TODO: measure the error an I-block encoding would do, and base the decision
- on that */
- swf_SetBits(tag, 1,1); /* cod=1, block skipped */
- /* copy the region from the last frame so that we have a complete reconstruction */
- copyregion(s, s->current, s->oldpic, bx, by);
- return 1;
+ memcpy(&fb_i, fb, sizeof(block_t));
+ dodctandquant(&fb_i, &data->b, 1, s->quant);
+ getblockpatterns(&data->b, &y, &c, 1);
+ *bits = 0;
+ if(!data->iframe) {
+ *bits += 1; //cod
}
+ *bits += data->ctable[c].len;
+ *bits += cbpy[y].len;
+ *bits += coefbits8x8(data->b.y1, 1);
+ *bits += coefbits8x8(data->b.y2, 1);
+ *bits += coefbits8x8(data->b.y3, 1);
+ *bits += coefbits8x8(data->b.y4, 1);
+ *bits += coefbits8x8(data->b.u, 1);
+ *bits += coefbits8x8(data->b.v, 1);
+ data->bits = *bits;
+
+ /* -- reconstruction -- */
+ memcpy(&data->reconstruction,&data->b,sizeof(block_t));
+ dequantize(&data->reconstruction, 1, s->quant);
+ doidct(&data->reconstruction);
+ truncateblock(&data->reconstruction);
+}
- predictmvd(s,bx,by,&predictmvdx,&predictmvdy);
- getregion(&fb, s->current, bx, by, s->width);
+int writeIBlock(VIDEOSTREAM*s, TAG*tag, iblockdata_t*data)
+{
+ int c = 0, y = 0;
+ int has_dc=1;
+ int bits = 0;
+ block_t b;
- { /* consider I-block */
- block_t fb_i;
- int y,c;
- memcpy(&fb_i, &fb, sizeof(block_t));
- dodctandquant(&fb_i, &b_i, 1, *quant);
- //quantize(&fb_i, &b_i, 1, *quant);
- getblockpatterns(&b_i, &y, &c, 1);
- bits_i = 1; //cod
- bits_i += mcbpc_inter[3*4+c].len;
- bits_i += cbpy[y].len;
- bits_i += coefbits8x8(b_i.y1, 1);
- bits_i += coefbits8x8(b_i.y2, 1);
- bits_i += coefbits8x8(b_i.y3, 1);
- bits_i += coefbits8x8(b_i.y4, 1);
- bits_i += coefbits8x8(b_i.u, 1);
- bits_i += coefbits8x8(b_i.v, 1);
+ getblockpatterns(&data->b, &y, &c, has_dc);
+ if(!data->iframe) {
+ swf_SetBits(tag,0,1); bits += 1; // COD
}
+ bits += codehuffman(tag, data->ctable, c);
+ bits += codehuffman(tag, cbpy, y);
- { /* consider mvd(x,y)-block */
- block_t fbdiff;
- int y,c;
+ /* luminance */
+ bits += encode8x8(tag, data->b.y1, has_dc, y&8);
+ bits += encode8x8(tag, data->b.y2, has_dc, y&4);
+ bits += encode8x8(tag, data->b.y3, has_dc, y&2);
+ bits += encode8x8(tag, data->b.y4, has_dc, y&1);
- x_v00=0;
- y_v00=0;
+ /* chrominance */
+ bits += encode8x8(tag, data->b.u, has_dc, c&2);
+ bits += encode8x8(tag, data->b.v, has_dc, c&1);
- if(s->do_motion) {
- int hx,hy;
- int bestx=0,besty=0,bestbits=65536;
- int startx=-2,endx=2;
- int starty=-2,endy=2;
+ copy_block_pic(s, s->current, &data->reconstruction, data->bx, data->by);
+ assert(data->bits == bits);
+ return bits;
+}
- if(!bx) startx=0;
- if(!by) starty=0;
- if(bx==s->bbx-1) endx=0;
- if(by==s->bby-1) endy=0;
+int getmvdbits(VIDEOSTREAM*s,block_t*fb, int bx,int by,int hx,int hy)
+{
+ block_t b;
+ block_t fbold;
+ block_t fbdiff;
+ int bits = 0;
+ memcpy(&fbdiff, fb, sizeof(block_t));
+ getmvdregion(&fbold, s->oldpic, bx, by, hx, hy, s->linex);
+ yuvdiff(&fbdiff, &fbold);
+ dodctandquant(&fbdiff, &b, 0, s->quant);
+ bits += coefbits8x8(b.y1, 0);
+ bits += coefbits8x8(b.y2, 0);
+ bits += coefbits8x8(b.y3, 0);
+ bits += coefbits8x8(b.y4, 0);
+ bits += coefbits8x8(b.u, 0);
+ bits += coefbits8x8(b.v, 0);
+ return bits;
+}
- for(hx=startx;hx<=endx;hx+=1)
- for(hy=starty;hy<=endy;hy+=1)
- {
- block_t b;
- block_t fbold;
- int bits = 0;
- memcpy(&fbdiff, &fb, sizeof(block_t));
- getmvdregion(&fbold, s->oldpic, bx, by, hx, hy, s->linex);
- yuvdiff(&fbdiff, &fbold);
- dodctandquant(&fbdiff, &b, 0, *quant);
- //quantize(&fbdiff, &b, 0, *quant);
- bits += coefbits8x8(b.y1, 0);
- bits += coefbits8x8(b.y2, 0);
- bits += coefbits8x8(b.y3, 0);
- bits += coefbits8x8(b.y4, 0);
- bits += coefbits8x8(b.u, 0);
- bits += coefbits8x8(b.v, 0);
- if(bits<bestbits) {
- bestbits = bits;
- bestx = hx;
- besty = hy;
- }
+void prepareMVDBlock(VIDEOSTREAM*s, mvdblockdata_t*data, int bx, int by, block_t* fb, int*bits)
+{ /* consider mvd(x,y)-block */
+
+ int t;
+ int y,c;
+ block_t fbdiff;
+ int predictmvdx;
+ int predictmvdy;
+
+ data->bx = bx;
+ data->by = by;
+ predictmvd(s,bx,by,&predictmvdx,&predictmvdy);
+
+ data->bits = 65535;
+ data->movex=0;
+ data->movey=0;
+
+ if(s->do_motion) {
+ int hx,hy;
+ int bestx=0,besty=0,bestbits=65536;
+ int startx=-32,endx=31;
+ int starty=-32,endy=31;
+
+ if(!bx) startx=0;
+ if(!by) starty=0;
+ if(bx==s->bbx-1) endx=0;
+ if(by==s->bby-1) endy=0;
+
+ for(hx=startx;hx<=endx;hx+=4)
+ for(hy=starty;hy<=endy;hy+=4)
+ {
+ int bits = 0;
+ bits = getmvdbits(s,fb,bx,by,hx,hy);
+ if(bits<bestbits) {
+ bestbits = bits;
+ bestx = hx;
+ besty = hy;
}
- x_v00 = bestx;
- y_v00 = besty;
}
-
- memcpy(&fbdiff, &fb, sizeof(block_t));
- getmvdregion(&fbold_v00, s->oldpic, bx, by, x_v00, y_v00, s->linex);
- yuvdiff(&fbdiff, &fbold_v00);
- dodctandquant(&fbdiff, &b_v00, 0, *quant);
- //quantize(&fbdiff, &b_v00, 0, *quant);
- getblockpatterns(&b_v00, &y, &c, 0);
-
- bits_v00 = 1; //cod
- bits_v00 += mcbpc_inter[0*4+c].len;
- bits_v00 += cbpy[y^15].len;
- bits_v00 += mvd[mvd2index(predictmvdx, predictmvdy, x_v00, y_v00, 0)].len; // (0,0)
- bits_v00 += mvd[mvd2index(predictmvdx, predictmvdy, x_v00, y_v00, 1)].len;
- bits_v00 += coefbits8x8(b_v00.y1, 0);
- bits_v00 += coefbits8x8(b_v00.y2, 0);
- bits_v00 += coefbits8x8(b_v00.y3, 0);
- bits_v00 += coefbits8x8(b_v00.y4, 0);
- bits_v00 += coefbits8x8(b_v00.u, 0);
- bits_v00 += coefbits8x8(b_v00.v, 0);
+
+ if(bestx-3 > startx) startx = bestx-3;
+ if(besty-3 > starty) starty = besty-3;
+ if(bestx+3 < endx) endx = bestx+3;
+ if(besty+3 < endy) endy = besty+3;
+
+ for(hx=startx;hx<=endx;hx++)
+ for(hy=starty;hy<=endy;hy++)
+ {
+ int bits = 0;
+ bits = getmvdbits(s,fb,bx,by,hx,hy);
+ if(bits<bestbits) {
+ bestbits = bits;
+ bestx = hx;
+ besty = hy;
+ }
+ }
+ data->movex = bestx;
+ data->movey = besty;
}
- if(bits_i > bits_v00)
- {
- /* mvd (0,0) block (mode=0) */
- int t;
- mode = 0; // mvd w/o mvd24
- has_dc = 0;
- memcpy(&b, &b_v00, sizeof(block_t));
-
- getblockpatterns(&b, &cbpybits, &cbpcbits, has_dc);
- swf_SetBits(tag,0,1); // COD
- codehuffman(tag, mcbpc_inter, mode*4+cbpcbits);
- codehuffman(tag, cbpy, cbpybits^15);
-
- /* vector */
- codehuffman(tag, mvd, mvd2index(predictmvdx, predictmvdy, x_v00, y_v00, 0));
- codehuffman(tag, mvd, mvd2index(predictmvdx, predictmvdy, x_v00, y_v00, 1));
- s->mvdx[by*s->bbx+bx] = x_v00;
- s->mvdy[by*s->bbx+bx] = y_v00;
-
- /* luminance */
- encode8x8(tag, b.y1, has_dc, cbpybits&8);
- encode8x8(tag, b.y2, has_dc, cbpybits&4);
- encode8x8(tag, b.y3, has_dc, cbpybits&2);
- encode8x8(tag, b.y4, has_dc, cbpybits&1);
-
- /* chrominance */
- encode8x8(tag, b.u, has_dc, cbpcbits&2);
- encode8x8(tag, b.v, has_dc, cbpcbits&1);
-
- /* -- reconstruction -- */
- dequantize(&b, 0, *quant);
- doidct(&b);
- for(t=0;t<64;t++) {
- b.y1[t] = truncate256(b.y1[t] + (int)fbold_v00.y1[t]);
- b.y2[t] = truncate256(b.y2[t] + (int)fbold_v00.y2[t]);
- b.y3[t] = truncate256(b.y3[t] + (int)fbold_v00.y3[t]);
- b.y4[t] = truncate256(b.y4[t] + (int)fbold_v00.y4[t]);
- b.u[t] = truncate256(b.u[t] + (int)fbold_v00.u[t]);
- b.v[t] = truncate256(b.v[t] + (int)fbold_v00.v[t]);
- }
- copyblock(s, s->current, &b, bx, by);
- return bits_v00;
- } else {
- /* i block (mode=3) */
- mode = 3;
- has_dc = 1;
- memcpy(&b, &b_i, sizeof(block_t));
- getblockpatterns(&b, &cbpybits, &cbpcbits, has_dc);
- swf_SetBits(tag,0,1); // COD
- codehuffman(tag, mcbpc_inter, mode*4+cbpcbits);
- codehuffman(tag, cbpy, cbpybits);
-
- /* luminance */
- encode8x8(tag, b.y1, has_dc, cbpybits&8);
- encode8x8(tag, b.y2, has_dc, cbpybits&4);
- encode8x8(tag, b.y3, has_dc, cbpybits&2);
- encode8x8(tag, b.y4, has_dc, cbpybits&1);
-
- /* chrominance */
- encode8x8(tag, b.u, has_dc, cbpcbits&2);
- encode8x8(tag, b.v, has_dc, cbpcbits&1);
-
- /* -- reconstruction -- */
- dequantize(&b, 1, *quant);
- doidct(&b);
- truncateblock(&b);
- copyblock(s, s->current, &b, bx, by);
- return bits_i;
+ memcpy(&fbdiff, fb, sizeof(block_t));
+ getmvdregion(&data->fbold, s->oldpic, bx, by, data->movex, data->movey, s->linex);
+ yuvdiff(&fbdiff, &data->fbold);
+ dodctandquant(&fbdiff, &data->b, 0, s->quant);
+ getblockpatterns(&data->b, &y, &c, 0);
+
+ data->xindex = mvd2index(predictmvdx, predictmvdy, data->movex, data->movey, 0);
+ data->yindex = mvd2index(predictmvdx, predictmvdy, data->movex, data->movey, 1);
+
+ *bits = 1; //cod
+ *bits += mcbpc_inter[0*4+c].len;
+ *bits += cbpy[y^15].len;
+ *bits += mvd[data->xindex].len; // (0,0)
+ *bits += mvd[data->yindex].len;
+ *bits += coefbits8x8(data->b.y1, 0);
+ *bits += coefbits8x8(data->b.y2, 0);
+ *bits += coefbits8x8(data->b.y3, 0);
+ *bits += coefbits8x8(data->b.y4, 0);
+ *bits += coefbits8x8(data->b.u, 0);
+ *bits += coefbits8x8(data->b.v, 0);
+ data->bits = *bits;
+
+ /* -- reconstruction -- */
+ memcpy(&data->reconstruction, &data->b, sizeof(block_t));
+ dequantize(&data->reconstruction, 0, s->quant);
+ doidct(&data->reconstruction);
+ for(t=0;t<64;t++) {
+ data->reconstruction.y1[t] =
+ truncate256(data->reconstruction.y1[t] + (int)data->fbold.y1[t]);
+ data->reconstruction.y2[t] =
+ truncate256(data->reconstruction.y2[t] + (int)data->fbold.y2[t]);
+ data->reconstruction.y3[t] =
+ truncate256(data->reconstruction.y3[t] + (int)data->fbold.y3[t]);
+ data->reconstruction.y4[t] =
+ truncate256(data->reconstruction.y4[t] + (int)data->fbold.y4[t]);
+ data->reconstruction.u[t] =
+ truncate256(data->reconstruction.u[t] + (int)data->fbold.u[t]);
+ data->reconstruction.v[t] =
+ truncate256(data->reconstruction.v[t] + (int)data->fbold.v[t]);
}
+}
- exit(1);
-#if 0
- dodct(&fb);
- quantize(&fb, &b, has_dc, *quant);
- getblockpatterns(&b, &cbpybits, &cbpcbits, has_dc);
+int writeMVDBlock(VIDEOSTREAM*s, TAG*tag, mvdblockdata_t*data)
+{
+ int c = 0, y = 0;
+ /* mvd (0,0) block (mode=0) */
+ int t;
+ int has_dc=0; // mvd w/o mvd24
+ int mode = 0;
+ int bx = data->bx;
+ int by = data->by;
+ int bits = 0;
- if(!dquant && has_mvd && !has_mvd24 && !has_dc) mode = 0;
- else if(dquant && has_mvd && !has_mvd24 && !has_dc) mode = 1;
- else if(!dquant && has_mvd && has_mvd24 && !has_dc) mode = 2;
- else if(!dquant && !has_mvd && !has_mvd24 && has_dc) mode = 3;
- else if(dquant && !has_mvd && !has_mvd24 && has_dc) mode = 4;
- else exit(1);
+ getblockpatterns(&data->b, &y, &c, has_dc);
+ swf_SetBits(tag,0,1); bits += 1; // COD
+ bits += codehuffman(tag, mcbpc_inter, mode*4+c);
+ bits += codehuffman(tag, cbpy, y^15);
- swf_SetBits(tag,0,1); /* cod - 1 if we're not going to code this block*/
+ /* vector */
+ bits += codehuffman(tag, mvd, data->xindex);
+ bits += codehuffman(tag, mvd, data->yindex);
- codehuffman(tag, mcbpc_inter, mode*4+cbpcbits);
- codehuffman(tag, cbpy, (mode==3 || mode==4)?cbpybits:cbpybits^15);
+ /* luminance */
+ bits += encode8x8(tag, data->b.y1, has_dc, y&8);
+ bits += encode8x8(tag, data->b.y2, has_dc, y&4);
+ bits += encode8x8(tag, data->b.y3, has_dc, y&2);
+ bits += encode8x8(tag, data->b.y4, has_dc, y&1);
- if(dquant) {
- setQuant(tag, dquant);
- }
+ /* chrominance */
+ bits += encode8x8(tag, data->b.u, has_dc, c&2);
+ bits += encode8x8(tag, data->b.v, has_dc, c&1);
+
+ s->mvdx[by*s->bbx+bx] = data->movex;
+ s->mvdy[by*s->bbx+bx] = data->movey;
+
+ copy_block_pic(s, s->current, &data->reconstruction, data->bx, data->by);
+ assert(data->bits == bits);
+ return bits;
+}
+
+static int encode_PFrame_block(TAG*tag, VIDEOSTREAM*s, int bx, int by)
+{
+ block_t fb;
+ int diff1,diff2;
+ int bits_i;
+ int bits_vxy;
- if(has_mvd) {
- /* 0,0 */
- codehuffman(tag, mvd, 32);
- codehuffman(tag, mvd, 32);
+ iblockdata_t iblock;
+ mvdblockdata_t mvdblock;
+
+ getregion(&fb, s->current, bx, by, s->linex);
+ prepareIBlock(s, &iblock, bx, by, &fb, &bits_i, 0);
+
+ /* encoded last frame <=> original current block: */
+ diff1 = compare_pic_pic(s, s->current, s->oldpic, bx, by);
+ /* encoded current frame <=> original current block: */
+ diff2 = compare_pic_block(s, &iblock.reconstruction, s->current, bx, by);
+
+ if(diff1 <= diff2) {
+ swf_SetBits(tag, 1,1); /* cod=1, block skipped */
+ /* copy the region from the last frame so that we have a complete reconstruction */
+ copyregion(s, s->current, s->oldpic, bx, by);
+ return 1;
}
- if(has_mvd24) {
+ prepareMVDBlock(s, &mvdblock, bx, by, &fb, &bits_vxy);
+
+ if(bits_i > bits_vxy) {
+ return writeMVDBlock(s, tag, &mvdblock);
+ } else {
+ return writeIBlock(s, tag, &iblock);
}
+}
- /* luminance */
- encode8x8(tag, b.y1, has_dc, cbpybits&8);
- encode8x8(tag, b.y2, has_dc, cbpybits&4);
- encode8x8(tag, b.y3, has_dc, cbpybits&2);
- encode8x8(tag, b.y4, has_dc, cbpybits&1);
+/* should be called encode_IFrameBlock */
+static void encode_IFrame_block(TAG*tag, VIDEOSTREAM*s, int bx, int by)
+{
+ block_t fb;
+ iblockdata_t data;
+ int bits;
- /* chrominance */
- encode8x8(tag, b.u, has_dc, cbpcbits&2);
- encode8x8(tag, b.v, has_dc, cbpcbits&1);
-#endif
+ getregion(&fb, s->current, bx, by, s->width);
+ prepareIBlock(s, &data, bx, by, &fb, &bits, 1);
+ writeIBlock(s, tag, &data);
}
+#ifdef MAIN
static int bmid = 0;
+
void setdbgpic(TAG*tag, RGBA*pic, int width, int height)
{
MATRIX m;
bmid++;
}
+#endif
#define TYPE_IFRAME 0
#define TYPE_PFRAME 1
swf_SetBits(tag, type, 2); /* I-Frame or P-Frame */
swf_SetBits(tag, 0, 1); /* No deblock filter */
assert(quant>0);
+ assert(quant<32);
swf_SetBits(tag, quant, 5); /* quantizer (1-31), may be updated later on*/
swf_SetBits(tag, 0, 1); /* No extra info */
}
if(quant<1) quant=1;
if(quant>31) quant=31;
+ s->quant = quant;
writeHeader(tag, s->width, s->height, s->frame, quant, TYPE_IFRAME);
- rgb2yuv(s->current, pic, s->linex, s->olinex, s->width, s->height);
+ /* fixme: should fill with 0,128,128, not 0,0,0 */
+ memset(s->current, 0, s->linex*s->height*sizeof(YUV));
- //dostat(s);
+ rgb2yuv(s->current, pic, s->linex, s->olinex, s->owidth, s->oheight);
for(by=0;by<s->bby;by++)
{
for(bx=0;bx<s->bbx;bx++)
{
- encode_blockI(tag, s, bx, by, &quant);
+ encode_IFrame_block(tag, s, bx, by);
}
}
s->frame++;
if(quant<1) quant=1;
if(quant>31) quant=31;
+ s->quant = quant;
writeHeader(tag, s->width, s->height, s->frame, quant, TYPE_PFRAME);
- rgb2yuv(s->current, pic, s->linex, s->olinex, s->width, s->height);
+ /* fixme: should fill with 0,128,128, not 0,0,0 */
+ memset(s->current, 0, s->linex*s->height*sizeof(YUV));
+
+ rgb2yuv(s->current, pic, s->linex, s->olinex, s->owidth, s->oheight);
memset(s->mvdx, 0, s->bbx*s->bby*sizeof(int));
memset(s->mvdy, 0, s->bbx*s->bby*sizeof(int));
{
for(bx=0;bx<s->bbx;bx++)
{
- encode_blockP(tag, s, bx, by, &quant);
+ encode_PFrame_block(tag, s, bx, by);
}
}
s->frame++;
memcpy(s->oldpic, s->current, s->width*s->height*sizeof(YUV));
-//#define PNG
#ifdef MAIN
#ifdef PNG
yuv2rgb(pic, s->current, s->linex, s->width, s->height);
setdbgpic(tag, pic, s->width, s->height);
#endif
- if(s->frame == (int)totalframes-1)
- {
- int t;
- FILE*fi = fopen("test.ppm", "wb");
- fprintf(fi, "P6\n%d %d\n255\n", s->width, s->height);
- for(t=0;t<s->width*s->height;t++)
- {
- fwrite(&pic[t].r, 1, 1, fi);
- fwrite(&pic[t].g, 1, 1, fi);
- fwrite(&pic[t].b, 1, 1, fi);
- }
- fclose(fi);
- }
#endif
}
-int uline[64],vline[64],yline[64];
+static int uline[64],vline[64],yline[64];
void swf_SetVideoStreamMover(TAG*tag, VIDEOSTREAM*s, int quant)
{
int bx, by;
int predictmvdx, predictmvdy;
//int mvx=-1+(2*(s->frame&1));
//int mvy=-1+((s->frame&2));
- int mvx=(lrand48()%4)-2;
+ int mvx=0;//(lrand48()%4)-2;
int mvy=3;
swf_SetBits(tag,0,1); // COD
}
}
+#define TESTS
+#ifdef TESTS
+void test_copy_diff()
+{
+ VIDEOSTREAM stream;
+ VIDEOSTREAM* s = &stream;
+ TAG*tag;
+ RGBA*pic = malloc(256*256*sizeof(RGBA));
+ block_t fb;
+ int x,y;
+ int bx,by;
+ for(x=0;x<256;x++)
+ for(y=0;y<256;y++) {
+ pic[y*256+x].r = x*y;
+ pic[y*256+x].g = x+y;
+ pic[y*256+x].b = (x+1)%(y+1);
+ }
+ tag = swf_InsertTag(0, ST_DEFINEVIDEOSTREAM);
+ swf_SetU16(tag, 33);
+ swf_SetVideoStreamDefine(tag, s, 10, 256, 256);
+
+ rgb2yuv(s->current, pic, s->linex, s->olinex, s->owidth, s->oheight);
+ for(by=0;by<16;by++)
+ for(bx=0;bx<16;bx++) {
+ int diff1,diff2;
+ /* test1: does compare pic pic return zero for identical blocks? */
+ diff1 = compare_pic_pic(s, s->current, s->current, bx, by);
+ assert(!diff1);
+ /* test2: do blocks which are copied back return zero diff? */
+ getregion(&fb, s->current, bx, by, s->linex);
+ copy_block_pic(s, s->oldpic, &fb, bx, by);
+ diff1 = compare_pic_block(s, &fb, s->oldpic, bx, by);
+ assert(!diff1);
+ /* test3: does compare_pic_block return the same result as compare_pic_pic? */
+ getregion(&fb, s->current, 15-bx, 15-by, s->linex);
+ copy_block_pic(s, s->oldpic, &fb, bx, by);
+ diff1 = compare_pic_block(s, &fb, s->current, bx, by);
+ diff2 = compare_pic_pic(s, s->current, s->oldpic, bx, by);
+ assert(diff1 == diff2);
+ }
+}
+
+#endif
+
#ifdef MAIN
#include "png.h"
int main(int argn, char*argv[])
SWFPLACEOBJECT obj;
int width = 0;
int height = 0;
- int frames = 20;
+ int frames = 10;
int framerate = 29;
unsigned char*data;
char* fname = "/home/kramm/pics/peppers.png";
VIDEOSTREAM stream;
double d = 1.0;
+#ifdef TESTS
+ test_copy_diff();
+#endif
+
memset(&stream, 0, sizeof(stream));
getPNG(fname, &width, &height, &data);
tag = swf_InsertTag(tag, ST_DEFINEVIDEOSTREAM);
swf_SetU16(tag, 33);
swf_SetVideoStreamDefine(tag, &stream, frames, width, height);
- stream.do_motion = 1;
+ stream.do_motion = 0;
for(t=0;t<frames;t++)
{
}
close(fi);
swf_FreeTags(&swf);
+ return 0;
}
#undef MAIN
#endif
git.asbjorn.biz / swftools.git / blobdiff