int main(int argc, char **argv) {
FILE *infile, *outfile[MAX_LAYERS];
vpx_codec_ctx_t codec;
vpx_codec_enc_cfg_t cfg;
int frame_cnt = 0;
vpx_image_t raw;
vpx_codec_err_t res;
unsigned int width;
unsigned int height;
int frame_avail;
int got_data;
int flags = 0;
int i;
int pts = 0; // PTS starts at 0
int frame_duration = 1; // 1 timebase tick per frame
int layering_mode = 0;
int frames_in_layer[MAX_LAYERS] = {0};
int layer_flags[MAX_PERIODICITY] = {0};
// Check usage and arguments
if (argc < 9)
die("Usage: %s <infile> <outfile> <width> <height> <rate_num> "
" <rate_den> <mode> <Rate_0> ... <Rate_nlayers-1>\n", argv[0]);
width = strtol (argv[3], NULL, 0);
height = strtol (argv[4], NULL, 0);
if (width < 16 || width%2 || height <16 || height%2)
die ("Invalid resolution: %d x %d", width, height);
if (!sscanf(argv[7], "%d", &layering_mode))
die ("Invalid mode %s", argv[7]);
if (layering_mode<0 || layering_mode>6)
die ("Invalid mode (0..6) %s", argv[7]);
if (argc != 8+mode_to_num_layers[layering_mode])
die ("Invalid number of arguments");
if (!vpx_img_alloc (&raw, VPX_IMG_FMT_I420, width, height, 1))
die ("Failed to allocate image", width, height);
printf("Using %s\n",vpx_codec_iface_name(interface));
// Populate encoder configuration
res = vpx_codec_enc_config_default(interface, &cfg, 0);
if(res) {
printf("Failed to get config: %s\n", vpx_codec_err_to_string(res));
return EXIT_FAILURE;
}
// Update the default configuration with our settings
cfg.g_w = width;
cfg.g_h = height;
// Timebase format e.g. 30fps: numerator=1, demoninator=30
if (!sscanf (argv[5], "%d", &cfg.g_timebase.num ))
die ("Invalid timebase numerator %s", argv[5]);
if (!sscanf (argv[6], "%d", &cfg.g_timebase.den ))
die ("Invalid timebase denominator %s", argv[6]);
for (i=8; i<8+mode_to_num_layers[layering_mode]; i++)
if (!sscanf(argv[i], "%d", &cfg.ts_target_bitrate[i-8]))
die ("Invalid data rate %s", argv[i]);
// Real time parameters
cfg.rc_dropframe_thresh = 0;
cfg.rc_end_usage = VPX_CBR;
cfg.rc_resize_allowed = 0;
cfg.rc_min_quantizer = 4;
cfg.rc_max_quantizer = 63;
cfg.rc_undershoot_pct = 98;
cfg.rc_overshoot_pct = 100;
cfg.rc_buf_initial_sz = 500;
cfg.rc_buf_optimal_sz = 600;
cfg.rc_buf_sz = 1000;
// Enable error resilient mode
cfg.g_error_resilient = 1;
cfg.g_lag_in_frames = 0;
cfg.kf_mode = VPX_KF_DISABLED;
// Disable automatic keyframe placement
cfg.kf_min_dist = cfg.kf_max_dist = 1000;
// Temporal scaling parameters:
// NOTE: The 3 prediction frames cannot be used interchangeably due to
// differences in the way they are handled throughout the code. The
// frames should be allocated to layers in the order LAST, GF, ARF.
// Other combinations work, but may produce slightly inferior results.
switch (layering_mode)
{
case 0:
{
// 2-layers, 2-frame period
int ids[2] = {0,1};
cfg.ts_number_layers = 2;
cfg.ts_periodicity = 2;
cfg.ts_rate_decimator[0] = 2;
cfg.ts_rate_decimator[1] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
#if 1
// 0=L, 1=GF, Intra-layer prediction enabled
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF;
layer_flags[1] = VP8_EFLAG_NO_UPD_ARF | VP8_EFLAG_NO_UPD_LAST |
VP8_EFLAG_NO_REF_ARF;
#else
// 0=L, 1=GF, Intra-layer prediction disabled
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF;
layer_flags[1] = VP8_EFLAG_NO_UPD_ARF | VP8_EFLAG_NO_UPD_LAST |
VP8_EFLAG_NO_REF_ARF | VP8_EFLAG_NO_REF_LAST;
#endif
break;
}
case 1:
{
// 2-layers, 3-frame period
int ids[3] = {0,1,1};
cfg.ts_number_layers = 2;
cfg.ts_periodicity = 3;
cfg.ts_rate_decimator[0] = 3;
cfg.ts_rate_decimator[1] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
// 0=L, 1=GF, Intra-layer prediction enabled
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF;
layer_flags[1] =
layer_flags[2] = VP8_EFLAG_NO_REF_GF |
VP8_EFLAG_NO_REF_ARF | VP8_EFLAG_NO_UPD_ARF |
VP8_EFLAG_NO_UPD_LAST;
break;
}
case 2:
{
// 3-layers, 6-frame period
int ids[6] = {0,2,2,1,2,2};
cfg.ts_number_layers = 3;
cfg.ts_periodicity = 6;
cfg.ts_rate_decimator[0] = 6;
cfg.ts_rate_decimator[1] = 3;
cfg.ts_rate_decimator[2] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
// 0=L, 1=GF, 2=ARF, Intra-layer prediction enabled
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF;
layer_flags[3] = VP8_EFLAG_NO_REF_ARF | VP8_EFLAG_NO_UPD_ARF |
VP8_EFLAG_NO_UPD_LAST;
layer_flags[1] =
layer_flags[2] =
layer_flags[4] =
layer_flags[5] = VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_LAST;
break;
}
case 3:
{
// 3-layers, 4-frame period
int ids[4] = {0,2,1,2};
cfg.ts_number_layers = 3;
cfg.ts_periodicity = 4;
cfg.ts_rate_decimator[0] = 4;
cfg.ts_rate_decimator[1] = 2;
cfg.ts_rate_decimator[2] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
// 0=L, 1=GF, 2=ARF, Intra-layer prediction disabled
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF;
layer_flags[2] = VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_ARF |
VP8_EFLAG_NO_UPD_LAST;
layer_flags[1] =
layer_flags[3] = VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_LAST | VP8_EFLAG_NO_UPD_GF |
VP8_EFLAG_NO_UPD_ARF;
break;
}
case 4:
{
// 3-layers, 4-frame period
int ids[4] = {0,2,1,2};
cfg.ts_number_layers = 3;
cfg.ts_periodicity = 4;
cfg.ts_rate_decimator[0] = 4;
cfg.ts_rate_decimator[1] = 2;
cfg.ts_rate_decimator[2] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
// 0=L, 1=GF, 2=ARF, Intra-layer prediction enabled in layer 1,
// disabled in layer 2
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF;
layer_flags[2] = VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_LAST | VP8_EFLAG_NO_UPD_ARF;
layer_flags[1] =
layer_flags[3] = VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_LAST | VP8_EFLAG_NO_UPD_GF |
VP8_EFLAG_NO_UPD_ARF;
break;
}
case 5:
{
// 3-layers, 4-frame period
int ids[4] = {0,2,1,2};
cfg.ts_number_layers = 3;
cfg.ts_periodicity = 4;
cfg.ts_rate_decimator[0] = 4;
cfg.ts_rate_decimator[1] = 2;
cfg.ts_rate_decimator[2] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
// 0=L, 1=GF, 2=ARF, Intra-layer prediction enabled
layer_flags[0] = VPX_EFLAG_FORCE_KF |
VP8_EFLAG_NO_REF_GF | VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_GF | VP8_EFLAG_NO_UPD_ARF;
layer_flags[2] = VP8_EFLAG_NO_REF_ARF |
VP8_EFLAG_NO_UPD_LAST | VP8_EFLAG_NO_UPD_ARF;
layer_flags[1] =
layer_flags[3] = VP8_EFLAG_NO_UPD_LAST | VP8_EFLAG_NO_UPD_GF;
break;
}
case 6:
{
// NOTE: Probably of academic interest only
// 5-layers, 16-frame period
int ids[16] = {0,4,3,4,2,4,3,4,1,4,3,4,2,4,3,4};
cfg.ts_number_layers = 5;
cfg.ts_periodicity = 16;
cfg.ts_rate_decimator[0] = 16;
cfg.ts_rate_decimator[1] = 8;
cfg.ts_rate_decimator[2] = 4;
cfg.ts_rate_decimator[3] = 2;
cfg.ts_rate_decimator[4] = 1;
memcpy(cfg.ts_layer_id, ids, sizeof(ids));
layer_flags[0] = VPX_EFLAG_FORCE_KF;
layer_flags[1] =
layer_flags[3] =
layer_flags[5] =
layer_flags[7] =
layer_flags[9] =
layer_flags[11] =
layer_flags[13] =
layer_flags[15] = VP8_EFLAG_NO_UPD_LAST |
VP8_EFLAG_NO_UPD_GF |
VP8_EFLAG_NO_UPD_ARF |
VP8_EFLAG_NO_UPD_ENTROPY;
layer_flags[2] =
layer_flags[6] =
layer_flags[10] =
layer_flags[14] = 0;
layer_flags[4] =
layer_flags[12] = VP8_EFLAG_NO_REF_LAST;
layer_flags[8] = VP8_EFLAG_NO_REF_LAST | VP8_EFLAG_NO_REF_GF |
VP8_EFLAG_NO_UPD_ENTROPY;
break;
}
default:
break;
}
// Open input file
if(!(infile = fopen(argv[1], "rb")))
die("Failed to open %s for reading", argv[1]);
// Open an output file for each stream
for (i=0; i<cfg.ts_number_layers; i++)
{
char file_name[512];
sprintf (file_name, "%s_%d.ivf", argv[2], i);
if (!(outfile[i] = fopen(file_name, "wb")))
die("Failed to open %s for writing", file_name);
write_ivf_file_header(outfile[i], &cfg, 0);
}
// Initialize codec
if (vpx_codec_enc_init (&codec, interface, &cfg, 0))
die_codec (&codec, "Failed to initialize encoder");
// Cap CPU & first I-frame size
vpx_codec_control (&codec, VP8E_SET_CPUUSED, -6);
vpx_codec_control (&codec, VP8E_SET_MAX_INTRA_BITRATE_PCT, 600);
frame_avail = 1;
while (frame_avail || got_data) {
vpx_codec_iter_t iter = NULL;
const vpx_codec_cx_pkt_t *pkt;
flags = layer_flags[frame_cnt % cfg.ts_periodicity];
frame_avail = read_frame(infile, &raw);
if (vpx_codec_encode(&codec, frame_avail? &raw : NULL, pts,
1, flags, VPX_DL_REALTIME))
die_codec(&codec, "Failed to encode frame");
// Reset KF flag
layer_flags[0] &= ~VPX_EFLAG_FORCE_KF;
got_data = 0;
while ( (pkt = vpx_codec_get_cx_data(&codec, &iter)) ) {
got_data = 1;
switch (pkt->kind) {
case VPX_CODEC_CX_FRAME_PKT:
for (i=cfg.ts_layer_id[frame_cnt % cfg.ts_periodicity];
i<cfg.ts_number_layers; i++)
{
write_ivf_frame_header(outfile[i], pkt);
if (fwrite(pkt->data.frame.buf, 1, pkt->data.frame.sz,
outfile[i]));
frames_in_layer[i]++;
}
break;
default:
break;
}
printf (pkt->kind == VPX_CODEC_CX_FRAME_PKT
&& (pkt->data.frame.flags & VPX_FRAME_IS_KEY)? "K":".");
fflush (stdout);
}
frame_cnt++;
pts += frame_duration;
}
printf ("\n");
fclose (infile);
printf ("Processed %d frames.\n",frame_cnt-1);
if (vpx_codec_destroy(&codec))
die_codec (&codec, "Failed to destroy codec");
// Try to rewrite the output file headers with the actual frame count
for (i=0; i<cfg.ts_number_layers; i++)
{
if (!fseek(outfile[i], 0, SEEK_SET))
write_ivf_file_header (outfile[i], &cfg, frames_in_layer[i]);
fclose (outfile[i]);
}
return EXIT_SUCCESS;
}
int main(int argc, char **argv) {
FILE *infile, *outfile;
vpx_codec_ctx_t codec;
vpx_codec_enc_cfg_t cfg;
int frame_cnt = 0;
vpx_image_t raw;
vpx_codec_err_t res;
long width;
long height;
int frame_avail;
int got_data;
int flags = 0;
/* Open files */
if(argc!=5)
die("Usage: %s <width> <height> <infile> <outfile>\n", argv[0]);
width = strtol(argv[1], NULL, 0);
height = strtol(argv[2], NULL, 0);
if(width < 16 || width%2 || height <16 || height%2)
die("Invalid resolution: %ldx%ld", width, height);
if(!vpx_img_alloc(&raw, VPX_IMG_FMT_I420, width, height, 1))
die("Faile to allocate image", width, height);
if(!(outfile = fopen(argv[4], "wb")))
die("Failed to open %s for writing", argv[4]);
printf("Using %s\n",vpx_codec_iface_name(interface));
/* Populate encoder configuration */ //
res = vpx_codec_enc_config_default(interface, &cfg, 0); //
if(res) { //
printf("Failed to get config: %s\n", vpx_codec_err_to_string(res)); //
return EXIT_FAILURE; //
} //
/* Update the default configuration with our settings */ //
cfg.rc_target_bitrate = width * height * cfg.rc_target_bitrate //
/ cfg.g_w / cfg.g_h; //
cfg.g_w = width; //
cfg.g_h = height; //
write_ivf_file_header(outfile, &cfg, 0);
/* Open input file for this encoding pass */
if(!(infile = fopen(argv[3], "rb")))
die("Failed to open %s for reading", argv[3]);
/* Initialize codec */ //
if(vpx_codec_enc_init(&codec, interface, &cfg, 0)) //
die_codec(&codec, "Failed to initialize encoder"); //
frame_avail = 1;
got_data = 0;
while(frame_avail || got_data) {
vpx_codec_iter_t iter = NULL;
const vpx_codec_cx_pkt_t *pkt;
frame_avail = read_frame(infile, &raw); //
if(vpx_codec_encode(&codec, frame_avail? &raw : NULL, frame_cnt, //
1, flags, VPX_DL_REALTIME)) //
die_codec(&codec, "Failed to encode frame"); //
got_data = 0;
while( (pkt = vpx_codec_get_cx_data(&codec, &iter)) ) {
got_data = 1;
switch(pkt->kind) {
case VPX_CODEC_CX_FRAME_PKT: //
write_ivf_frame_header(outfile, pkt); //
if(fwrite(pkt->data.frame.buf, 1, pkt->data.frame.sz, //
outfile)); //
break; //
default:
break;
}
printf(pkt->kind == VPX_CODEC_CX_FRAME_PKT
&& (pkt->data.frame.flags & VPX_FRAME_IS_KEY)? "K":".");
fflush(stdout);
}
frame_cnt++;
}
printf("\n");
fclose(infile);
printf("Processed %d frames.\n",frame_cnt-1);
if(vpx_codec_destroy(&codec)) //
die_codec(&codec, "Failed to destroy codec"); //
/* Try to rewrite the file header with the actual frame count */
if(!fseek(outfile, 0, SEEK_SET))
write_ivf_file_header(outfile, &cfg, frame_cnt-1);
fclose(outfile);
return EXIT_SUCCESS;
}
int main(int argc, char **argv) {
FILE *infile, *outfile;
vpx_codec_ctx_t codec;
vpx_codec_enc_cfg_t cfg;
int frame_cnt = 0;
vpx_image_t raw;
vpx_codec_err_t res;
long width;
long height;
int frame_avail;
int got_data;
int flags = 0;
/* Open files */
if(argc!=5)
die("Usage: %s <width> <height> <infile> <outfile>\n", argv[0]);
width = strtol(argv[1], NULL, 0);
height = strtol(argv[2], NULL, 0);
if(width < 16 || width%2 || height <16 || height%2)
die("Invalid resolution: %ldx%ld", width, height);
if(!vpx_img_alloc(&raw, VPX_IMG_FMT_I420, width, height, 1))
die("Faile to allocate image", width, height);
if(!(outfile = fopen(argv[4], "wb")))
die("Failed to open %s for writing", argv[4]);
printf("Using %s\n",vpx_codec_iface_name(interface));
/* Populate encoder configuration */
res = vpx_codec_enc_config_default(interface, &cfg, 0);
if(res) {
printf("Failed to get config: %s\n", vpx_codec_err_to_string(res));
return EXIT_FAILURE;
}
/* Update the default configuration with our settings */
cfg.rc_target_bitrate = width * height * cfg.rc_target_bitrate
/ cfg.g_w / cfg.g_h;
cfg.g_w = width;
cfg.g_h = height;
write_ivf_file_header(outfile, &cfg, 0);
/* Open input file for this encoding pass */
if(!(infile = fopen(argv[3], "rb")))
die("Failed to open %s for reading", argv[3]);
/* Initialize codec */
if(vpx_codec_enc_init(&codec, interface, &cfg, 0))
die_codec(&codec, "Failed to initialize encoder");
frame_avail = 1;
got_data = 0;
while(frame_avail || got_data) {
vpx_codec_iter_t iter = NULL;
const vpx_codec_cx_pkt_t *pkt;
if(frame_cnt + 1 == 22) { //
vpx_roi_map_t roi; //
int i; //
//
roi.rows = cfg.g_h/16; //
roi.cols = cfg.g_w/16; //
//
roi.delta_q[0] = 0; //
roi.delta_q[1] = -2; //
roi.delta_q[2] = -4; //
roi.delta_q[3] = -6; //
//
roi.delta_lf[0] = 0; //
roi.delta_lf[1] = 1; //
roi.delta_lf[2] = 2; //
roi.delta_lf[3] = 3; //
//
roi.static_threshold[0] = 1500; //
roi.static_threshold[1] = 1000; //
roi.static_threshold[2] = 500; //
roi.static_threshold[3] = 0; //
//
/* generate an ROI map for example */ //
roi.roi_map = malloc(roi.rows * roi.cols); //
for(i=0;i<roi.rows*roi.cols;i++) //
roi.roi_map[i] = i & 3; //
//
if(vpx_codec_control(&codec, VP8E_SET_ROI_MAP, &roi)) //
die_codec(&codec, "Failed to set ROI map"); //
//
free(roi.roi_map); //
} else if(frame_cnt + 1 == 33) { //
vpx_active_map_t active; //
int i; //
//
active.rows = cfg.g_h/16; //
active.cols = cfg.g_w/16; //
//
/* generate active map for example */ //
active.active_map = malloc(active.rows * active.cols); //
for(i=0;i<active.rows*active.cols;i++) //
active.active_map[i] = i & 1; //
//
if(vpx_codec_control(&codec, VP8E_SET_ACTIVEMAP, &active)) //
die_codec(&codec, "Failed to set active map"); //
//
free(active.active_map); //
} else if(frame_cnt + 1 == 44) { //
vpx_active_map_t active; //
//
active.rows = cfg.g_h/16; //
active.cols = cfg.g_w/16; //
//
/* pass in null map to disable active_map*/ //
active.active_map = NULL; //
//
if(vpx_codec_control(&codec, VP8E_SET_ACTIVEMAP, &active)) //
die_codec(&codec, "Failed to set active map"); //
} //
frame_avail = read_frame(infile, &raw);
if(vpx_codec_encode(&codec, frame_avail? &raw : NULL, frame_cnt,
1, flags, VPX_DL_REALTIME))
die_codec(&codec, "Failed to encode frame");
got_data = 0;
while( (pkt = vpx_codec_get_cx_data(&codec, &iter)) ) {
got_data = 1;
switch(pkt->kind) {
case VPX_CODEC_CX_FRAME_PKT:
write_ivf_frame_header(outfile, pkt);
if(fwrite(pkt->data.frame.buf, 1, pkt->data.frame.sz,
outfile));
break;
default:
break;
}
printf(pkt->kind == VPX_CODEC_CX_FRAME_PKT
&& (pkt->data.frame.flags & VPX_FRAME_IS_KEY)? "K":".");
fflush(stdout);
}
frame_cnt++;
}
printf("\n");
fclose(infile);
printf("Processed %d frames.\n",frame_cnt-1);
if(vpx_codec_destroy(&codec))
die_codec(&codec, "Failed to destroy codec");
/* Try to rewrite the file header with the actual frame count */
if(!fseek(outfile, 0, SEEK_SET))
write_ivf_file_header(outfile, &cfg, frame_cnt-1);
fclose(outfile);
return EXIT_SUCCESS;
}
int main(int argc, char **argv) {
FILE *infile, *outfile;
vpx_codec_ctx_t codec;
vpx_codec_enc_cfg_t cfg;
int frame_cnt = 0;
unsigned char file_hdr[IVF_FILE_HDR_SZ];
unsigned char frame_hdr[IVF_FRAME_HDR_SZ];
vpx_image_t raw;
vpx_codec_err_t ret;
int width,height;
int y_size;
int frame_avail;
int got_data;
int flags = 0;
width = 640;
height = 360;
infile = fopen("../cuc_ieschool_640x360_yuv420p.yuv", "rb");
outfile = fopen("cuc_ieschool.ivf", "wb");
if(infile==NULL||outfile==NULL){
printf("Error open files.\n");
return -1;
}
if(!vpx_img_alloc(&raw, VPX_IMG_FMT_I420, width, height, 1)){
printf("Fail to allocate image\n");
return -1;
}
printf("Using %s\n",vpx_codec_iface_name(interface));
//Populate encoder configuration
ret = vpx_codec_enc_config_default(interface, &cfg, 0);
if(ret) {
printf("Failed to get config: %s\n", vpx_codec_err_to_string(ret));
return -1;
}
cfg.rc_target_bitrate =800;
cfg.g_w = width;
cfg.g_h = height;
write_ivf_file_header(outfile, &cfg, 0);
//Initialize codec
if(vpx_codec_enc_init(&codec, interface, &cfg, 0)){
printf("Failed to initialize encoder\n");
return -1;
}
frame_avail = 1;
got_data = 0;
y_size=cfg.g_w*cfg.g_h;
while(frame_avail || got_data) {
vpx_codec_iter_t iter = NULL;
const vpx_codec_cx_pkt_t *pkt;
if(fread(raw.planes[0], 1, y_size*3/2, infile)!=y_size*3/2){
frame_avail=0;
}
if(frame_avail){
//Encode
ret=vpx_codec_encode(&codec,&raw,frame_cnt,1,flags,VPX_DL_REALTIME);
}else{
//Flush Encoder
ret=vpx_codec_encode(&codec,NULL,frame_cnt,1,flags,VPX_DL_REALTIME);
}
if(ret){
printf("Failed to encode frame\n");
return -1;
}
got_data = 0;
while( (pkt = vpx_codec_get_cx_data(&codec, &iter)) ) {
got_data = 1;
switch(pkt->kind) {
case VPX_CODEC_CX_FRAME_PKT:
write_ivf_frame_header(outfile, pkt);
fwrite(pkt->data.frame.buf, 1, pkt->data.frame.sz,outfile);
break;
default:
break;
}
}
printf("Succeed encode frame: %5d\n",frame_cnt);
frame_cnt++;
}
fclose(infile);
vpx_codec_destroy(&codec);
//Try to rewrite the file header with the actual frame count
if(!fseek(outfile, 0, SEEK_SET))
write_ivf_file_header(outfile, &cfg, frame_cnt-1);
fclose(outfile);
return 0;
}
int main(int argc, char **argv)
{
FILE *infile, *outfile[NUM_ENCODERS];
vpx_codec_ctx_t codec[NUM_ENCODERS];
vpx_codec_enc_cfg_t cfg[NUM_ENCODERS];
vpx_codec_pts_t frame_cnt = 0;
vpx_image_t raw[NUM_ENCODERS];
vpx_codec_err_t res[NUM_ENCODERS];
int i;
long width;
long height;
int frame_avail;
int got_data;
int flags = 0;
/*Currently, only realtime mode is supported in multi-resolution encoding.*/
int arg_deadline = VPX_DL_REALTIME;
/* Set show_psnr to 1/0 to show/not show PSNR. Choose show_psnr=0 if you
don't need to know PSNR, which will skip PSNR calculation and save
encoding time. */
int show_psnr = 0;
uint64_t psnr_sse_total[NUM_ENCODERS] = {0};
uint64_t psnr_samples_total[NUM_ENCODERS] = {0};
double psnr_totals[NUM_ENCODERS][4] = {{0,0}};
int psnr_count[NUM_ENCODERS] = {0};
/* Set the required target bitrates for each resolution level. */
unsigned int target_bitrate[NUM_ENCODERS]={1400, 500, 100};
/* Enter the frame rate of the input video */
int framerate = 30;
/* Set down-sampling factor for each resolution level.
dsf[0] controls down sampling from level 0 to level 1;
dsf[1] controls down sampling from level 1 to level 2;
dsf[2] is not used. */
vpx_rational_t dsf[NUM_ENCODERS] = {{2, 1}, {2, 1}, {1, 1}};
if(argc!= (5+NUM_ENCODERS))
die("Usage: %s <width> <height> <infile> <outfile(s)> <output psnr?>\n",
argv[0]);
printf("Using %s\n",vpx_codec_iface_name(interface));
width = strtol(argv[1], NULL, 0);
height = strtol(argv[2], NULL, 0);
if(width < 16 || width%2 || height <16 || height%2)
die("Invalid resolution: %ldx%ld", width, height);
/* Open input video file for encoding */
if(!(infile = fopen(argv[3], "rb")))
die("Failed to open %s for reading", argv[3]);
/* Open output file for each encoder to output bitstreams */
for (i=0; i< NUM_ENCODERS; i++)
{
if(!(outfile[i] = fopen(argv[i+4], "wb")))
die("Failed to open %s for writing", argv[i+4]);
}
show_psnr = strtol(argv[NUM_ENCODERS + 4], NULL, 0);
/* Populate default encoder configuration */
for (i=0; i< NUM_ENCODERS; i++)
{
res[i] = vpx_codec_enc_config_default(interface, &cfg[i], 0);
if(res[i]) {
printf("Failed to get config: %s\n", vpx_codec_err_to_string(res[i]));
return EXIT_FAILURE;
}
}
/*
* Update the default configuration according to needs of the application.
*/
/* Highest-resolution encoder settings */
cfg[0].g_w = width;
cfg[0].g_h = height;
cfg[0].g_threads = 1; /* number of threads used */
cfg[0].rc_dropframe_thresh = 0;
cfg[0].rc_end_usage = VPX_CBR;
cfg[0].rc_resize_allowed = 0;
cfg[0].rc_min_quantizer = 4;
cfg[0].rc_max_quantizer = 56;
cfg[0].rc_undershoot_pct = 98;
cfg[0].rc_overshoot_pct = 100;
cfg[0].rc_buf_initial_sz = 500;
cfg[0].rc_buf_optimal_sz = 600;
cfg[0].rc_buf_sz = 1000;
//cfg[0].rc_dropframe_thresh = 10;
cfg[0].g_error_resilient = 1; /* Enable error resilient mode */
cfg[0].g_lag_in_frames = 0;
/* Disable automatic keyframe placement */
//cfg[0].kf_mode = VPX_KF_DISABLED;
cfg[0].kf_min_dist = cfg[0].kf_max_dist = 1000;
cfg[0].rc_target_bitrate = target_bitrate[0]; /* Set target bitrate */
cfg[0].g_timebase.num = 1; /* Set fps */
cfg[0].g_timebase.den = framerate;
/* Other-resolution encoder settings */
for (i=1; i< NUM_ENCODERS; i++)
{
memcpy(&cfg[i], &cfg[0], sizeof(vpx_codec_enc_cfg_t));
cfg[i].g_threads = 1; /* number of threads used */
cfg[i].rc_target_bitrate = target_bitrate[i];
/* Note: Width & height of other-resolution encoders are calculated
* from the highest-resolution encoder's size and the corresponding
* down_sampling_factor.
*/
{
unsigned int iw = cfg[i-1].g_w*dsf[i-1].den + dsf[i-1].num - 1;
unsigned int ih = cfg[i-1].g_h*dsf[i-1].den + dsf[i-1].num - 1;
cfg[i].g_w = iw/dsf[i-1].num;
cfg[i].g_h = ih/dsf[i-1].num;
}
/* Make width & height to be multiplier of 2. */
// Should support odd size ???
if((cfg[i].g_w)%2)cfg[i].g_w++;
if((cfg[i].g_h)%2)cfg[i].g_h++;
}
/* Allocate image for each encoder */
for (i=0; i< NUM_ENCODERS; i++)
if(!vpx_img_alloc(&raw[i], VPX_IMG_FMT_I420, cfg[i].g_w, cfg[i].g_h, 32))
die("Failed to allocate image", cfg[i].g_w, cfg[i].g_h);
if (raw[0].stride[VPX_PLANE_Y] == raw[0].d_w)
read_frame_p = read_frame;
else
read_frame_p = read_frame_by_row;
for (i=0; i< NUM_ENCODERS; i++)
write_ivf_file_header(outfile[i], &cfg[i], 0);
/* Initialize multi-encoder */
if(vpx_codec_enc_init_multi(&codec[0], interface, &cfg[0], NUM_ENCODERS,
(show_psnr ? VPX_CODEC_USE_PSNR : 0), &dsf[0]))
die_codec(&codec[0], "Failed to initialize encoder");
/* The extra encoding configuration parameters can be set as follows. */
/* Set encoding speed */
for ( i=0; i<NUM_ENCODERS; i++)
{
int speed = -6;
if(vpx_codec_control(&codec[i], VP8E_SET_CPUUSED, speed))
die_codec(&codec[i], "Failed to set cpu_used");
}
/* Set static thresh for highest-resolution encoder. Set it to 1000 for
* better performance. */
{
unsigned int static_thresh = 1000;
if(vpx_codec_control(&codec[0], VP8E_SET_STATIC_THRESHOLD, static_thresh))
die_codec(&codec[0], "Failed to set static threshold");
}
/* Set static thresh = 0 for other encoders for better quality */
for ( i=1; i<NUM_ENCODERS; i++)
{
unsigned int static_thresh = 0;
if(vpx_codec_control(&codec[i], VP8E_SET_STATIC_THRESHOLD, static_thresh))
die_codec(&codec[i], "Failed to set static threshold");
}
frame_avail = 1;
got_data = 0;
while(frame_avail || got_data)
{
vpx_codec_iter_t iter[NUM_ENCODERS]={NULL};
const vpx_codec_cx_pkt_t *pkt[NUM_ENCODERS];
flags = 0;
frame_avail = read_frame_p(infile, &raw[0]);
if(frame_avail)
{
for ( i=1; i<NUM_ENCODERS; i++)
{
/*Scale the image down a number of times by downsampling factor*/
/* FilterMode 1 or 2 give better psnr than FilterMode 0. */
I420Scale(raw[i-1].planes[VPX_PLANE_Y], raw[i-1].stride[VPX_PLANE_Y],
raw[i-1].planes[VPX_PLANE_U], raw[i-1].stride[VPX_PLANE_U],
raw[i-1].planes[VPX_PLANE_V], raw[i-1].stride[VPX_PLANE_V],
raw[i-1].d_w, raw[i-1].d_h,
raw[i].planes[VPX_PLANE_Y], raw[i].stride[VPX_PLANE_Y],
raw[i].planes[VPX_PLANE_U], raw[i].stride[VPX_PLANE_U],
raw[i].planes[VPX_PLANE_V], raw[i].stride[VPX_PLANE_V],
raw[i].d_w, raw[i].d_h, 1);
}
}
/* Encode each frame at multi-levels */
if(vpx_codec_encode(&codec[0], frame_avail? &raw[0] : NULL,
frame_cnt, 1, flags, arg_deadline))
die_codec(&codec[0], "Failed to encode frame");
for (i=NUM_ENCODERS-1; i>=0 ; i--)
{
got_data = 0;
while( (pkt[i] = vpx_codec_get_cx_data(&codec[i], &iter[i])) )
{
got_data = 1;
switch(pkt[i]->kind) {
case VPX_CODEC_CX_FRAME_PKT:
write_ivf_frame_header(outfile[i], pkt[i]);
if(fwrite(pkt[i]->data.frame.buf, 1, pkt[i]->data.frame.sz,
outfile[i]));
break;
case VPX_CODEC_PSNR_PKT:
if (show_psnr)
{
int j;
psnr_sse_total[i] += pkt[i]->data.psnr.sse[0];
psnr_samples_total[i] += pkt[i]->data.psnr.samples[0];
for (j = 0; j < 4; j++)
{
//fprintf(stderr, "%.3lf ", pkt[i]->data.psnr.psnr[j]);
psnr_totals[i][j] += pkt[i]->data.psnr.psnr[j];
}
psnr_count[i]++;
}
break;
default:
break;
}
printf(pkt[i]->kind == VPX_CODEC_CX_FRAME_PKT
&& (pkt[i]->data.frame.flags & VPX_FRAME_IS_KEY)? "K":".");
fflush(stdout);
}
}
frame_cnt++;
}
printf("\n");
fclose(infile);
for (i=0; i< NUM_ENCODERS; i++)
{
printf("Processed %ld frames.\n",(long int)frame_cnt-1);
/* Calculate PSNR and print it out */
if ( (show_psnr) && (psnr_count[i]>0) )
{
int j;
double ovpsnr = vp8_mse2psnr(psnr_samples_total[i], 255.0,
psnr_sse_total[i]);
fprintf(stderr, "\n ENC%d PSNR (Overall/Avg/Y/U/V)", i);
fprintf(stderr, " %.3lf", ovpsnr);
for (j = 0; j < 4; j++)
{
fprintf(stderr, " %.3lf", psnr_totals[i][j]/psnr_count[i]);
}
}
if(vpx_codec_destroy(&codec[i]))
die_codec(&codec[i], "Failed to destroy codec");
/* Try to rewrite the file header with the actual frame count */
if(!fseek(outfile[i], 0, SEEK_SET))
write_ivf_file_header(outfile[i], &cfg[i], frame_cnt-1);
fclose(outfile[i]);
vpx_img_free(&raw[i]);
}
return EXIT_SUCCESS;
}
int main(int argc, const char **argv_)
{
vpx_codec_ctx_t encoder;
const char *in_fn = NULL, *out_fn = NULL, *stats_fn = NULL;
int i;
FILE *infile, *outfile;
vpx_codec_enc_cfg_t cfg;
vpx_codec_err_t res;
int pass, one_pass_only = 0;
stats_io_t stats;
vpx_image_t raw;
const struct codec_item *codec = codecs;
int frame_avail, got_data;
struct arg arg;
char **argv, **argi, **argj;
int arg_usage = 0, arg_passes = 1, arg_deadline = 0;
int arg_ctrls[ARG_CTRL_CNT_MAX][2], arg_ctrl_cnt = 0;
int arg_limit = 0;
static const arg_def_t **ctrl_args = no_args;
static const int *ctrl_args_map = NULL;
int verbose = 0, show_psnr = 0;
int arg_use_i420 = 1;
int arg_have_timebase = 0;
unsigned long cx_time = 0;
unsigned int file_type, fourcc;
y4m_input y4m;
exec_name = argv_[0];
if (argc < 3)
usage_exit();
/* First parse the codec and usage values, because we want to apply other
* parameters on top of the default configuration provided by the codec.
*/
argv = argv_dup(argc - 1, argv_ + 1);
for (argi = argj = argv; (*argj = *argi); argi += arg.argv_step)
{
arg.argv_step = 1;
if (arg_match(&arg, &codecarg, argi))
{
int j, k = -1;
for (j = 0; j < sizeof(codecs) / sizeof(codecs[0]); j++)
if (!strcmp(codecs[j].name, arg.val))
k = j;
if (k >= 0)
codec = codecs + k;
else
die("Error: Unrecognized argument (%s) to --codec\n",
arg.val);
}
else if (arg_match(&arg, &passes, argi))
{
arg_passes = arg_parse_uint(&arg);
if (arg_passes < 1 || arg_passes > 2)
die("Error: Invalid number of passes (%d)\n", arg_passes);
}
else if (arg_match(&arg, &pass_arg, argi))
{
one_pass_only = arg_parse_uint(&arg);
if (one_pass_only < 1 || one_pass_only > 2)
die("Error: Invalid pass selected (%d)\n", one_pass_only);
}
else if (arg_match(&arg, &fpf_name, argi))
stats_fn = arg.val;
else if (arg_match(&arg, &usage, argi))
arg_usage = arg_parse_uint(&arg);
else if (arg_match(&arg, &deadline, argi))
arg_deadline = arg_parse_uint(&arg);
else if (arg_match(&arg, &best_dl, argi))
arg_deadline = VPX_DL_BEST_QUALITY;
else if (arg_match(&arg, &good_dl, argi))
arg_deadline = VPX_DL_GOOD_QUALITY;
else if (arg_match(&arg, &rt_dl, argi))
arg_deadline = VPX_DL_REALTIME;
else if (arg_match(&arg, &use_yv12, argi))
{
arg_use_i420 = 0;
}
else if (arg_match(&arg, &use_i420, argi))
{
arg_use_i420 = 1;
}
else if (arg_match(&arg, &verbosearg, argi))
verbose = 1;
else if (arg_match(&arg, &limit, argi))
arg_limit = arg_parse_uint(&arg);
else if (arg_match(&arg, &psnrarg, argi))
show_psnr = 1;
else
argj++;
}
/* Ensure that --passes and --pass are consistent. If --pass is set and --passes=2,
* ensure --fpf was set.
*/
if (one_pass_only)
{
/* DWIM: Assume the user meant passes=2 if pass=2 is specified */
if (one_pass_only > arg_passes)
{
fprintf(stderr, "Warning: Assuming --pass=%d implies --passes=%d\n",
one_pass_only, one_pass_only);
arg_passes = one_pass_only;
}
if (arg_passes == 2 && !stats_fn)
die("Must specify --fpf when --pass=%d and --passes=2\n", one_pass_only);
}
/* Populate encoder configuration */
res = vpx_codec_enc_config_default(codec->iface, &cfg, arg_usage);
if (res)
{
fprintf(stderr, "Failed to get config: %s\n",
vpx_codec_err_to_string(res));
return EXIT_FAILURE;
}
/* Now parse the remainder of the parameters. */
for (argi = argj = argv; (*argj = *argi); argi += arg.argv_step)
{
arg.argv_step = 1;
if (0);
else if (arg_match(&arg, &threads, argi))
cfg.g_threads = arg_parse_uint(&arg);
else if (arg_match(&arg, &profile, argi))
cfg.g_profile = arg_parse_uint(&arg);
else if (arg_match(&arg, &width, argi))
cfg.g_w = arg_parse_uint(&arg);
else if (arg_match(&arg, &height, argi))
cfg.g_h = arg_parse_uint(&arg);
else if (arg_match(&arg, &timebase, argi))
{
cfg.g_timebase = arg_parse_rational(&arg);
arg_have_timebase = 1;
}
else if (arg_match(&arg, &error_resilient, argi))
cfg.g_error_resilient = arg_parse_uint(&arg);
else if (arg_match(&arg, &lag_in_frames, argi))
cfg.g_lag_in_frames = arg_parse_uint(&arg);
else if (arg_match(&arg, &dropframe_thresh, argi))
cfg.rc_dropframe_thresh = arg_parse_uint(&arg);
else if (arg_match(&arg, &resize_allowed, argi))
cfg.rc_resize_allowed = arg_parse_uint(&arg);
else if (arg_match(&arg, &resize_up_thresh, argi))
cfg.rc_resize_up_thresh = arg_parse_uint(&arg);
else if (arg_match(&arg, &resize_down_thresh, argi))
cfg.rc_resize_down_thresh = arg_parse_uint(&arg);
else if (arg_match(&arg, &resize_down_thresh, argi))
cfg.rc_resize_down_thresh = arg_parse_uint(&arg);
else if (arg_match(&arg, &end_usage, argi))
cfg.rc_end_usage = arg_parse_uint(&arg);
else if (arg_match(&arg, &target_bitrate, argi))
cfg.rc_target_bitrate = arg_parse_uint(&arg);
else if (arg_match(&arg, &min_quantizer, argi))
cfg.rc_min_quantizer = arg_parse_uint(&arg);
else if (arg_match(&arg, &max_quantizer, argi))
cfg.rc_max_quantizer = arg_parse_uint(&arg);
else if (arg_match(&arg, &undershoot_pct, argi))
cfg.rc_undershoot_pct = arg_parse_uint(&arg);
else if (arg_match(&arg, &overshoot_pct, argi))
cfg.rc_overshoot_pct = arg_parse_uint(&arg);
else if (arg_match(&arg, &buf_sz, argi))
cfg.rc_buf_sz = arg_parse_uint(&arg);
else if (arg_match(&arg, &buf_initial_sz, argi))
cfg.rc_buf_initial_sz = arg_parse_uint(&arg);
else if (arg_match(&arg, &buf_optimal_sz, argi))
cfg.rc_buf_optimal_sz = arg_parse_uint(&arg);
else if (arg_match(&arg, &bias_pct, argi))
{
cfg.rc_2pass_vbr_bias_pct = arg_parse_uint(&arg);
if (arg_passes < 2)
fprintf(stderr,
"Warning: option %s ignored in one-pass mode.\n",
arg.name);
}
else if (arg_match(&arg, &minsection_pct, argi))
{
cfg.rc_2pass_vbr_minsection_pct = arg_parse_uint(&arg);
if (arg_passes < 2)
fprintf(stderr,
"Warning: option %s ignored in one-pass mode.\n",
arg.name);
}
else if (arg_match(&arg, &maxsection_pct, argi))
{
cfg.rc_2pass_vbr_maxsection_pct = arg_parse_uint(&arg);
if (arg_passes < 2)
fprintf(stderr,
"Warning: option %s ignored in one-pass mode.\n",
arg.name);
}
else if (arg_match(&arg, &kf_min_dist, argi))
cfg.kf_min_dist = arg_parse_uint(&arg);
else if (arg_match(&arg, &kf_max_dist, argi))
cfg.kf_max_dist = arg_parse_uint(&arg);
else if (arg_match(&arg, &kf_disabled, argi))
cfg.kf_mode = VPX_KF_DISABLED;
else
argj++;
}
/* Handle codec specific options */
#if CONFIG_VP8_ENCODER
if (codec->iface == &vpx_codec_vp8_cx_algo)
{
ctrl_args = vp8_args;
ctrl_args_map = vp8_arg_ctrl_map;
}
#endif
for (argi = argj = argv; (*argj = *argi); argi += arg.argv_step)
{
int match = 0;
arg.argv_step = 1;
for (i = 0; ctrl_args[i]; i++)
{
if (arg_match(&arg, ctrl_args[i], argi))
{
match = 1;
if (arg_ctrl_cnt < ARG_CTRL_CNT_MAX)
{
arg_ctrls[arg_ctrl_cnt][0] = ctrl_args_map[i];
arg_ctrls[arg_ctrl_cnt][1] = arg_parse_int(&arg);
arg_ctrl_cnt++;
}
}
}
if (!match)
argj++;
}
/* Check for unrecognized options */
for (argi = argv; *argi; argi++)
if (argi[0][0] == '-' && argi[0][1])
die("Error: Unrecognized option %s\n", *argi);
/* Handle non-option arguments */
in_fn = argv[0];
out_fn = argv[1];
if (!in_fn || !out_fn)
usage_exit();
memset(&stats, 0, sizeof(stats));
for (pass = one_pass_only ? one_pass_only - 1 : 0; pass < arg_passes; pass++)
{
int frames_in = 0, frames_out = 0;
unsigned long nbytes = 0;
struct detect_buffer detect;
/* Parse certain options from the input file, if possible */
infile = strcmp(in_fn, "-") ? fopen(in_fn, "rb") : stdin;
if (!infile)
{
fprintf(stderr, "Failed to open input file\n");
return EXIT_FAILURE;
}
fread(detect.buf, 1, 4, infile);
detect.valid = 0;
if (file_is_y4m(infile, &y4m, detect.buf))
{
if (y4m_input_open(&y4m, infile, detect.buf, 4) >= 0)
{
file_type = FILE_TYPE_Y4M;
cfg.g_w = y4m.pic_w;
cfg.g_h = y4m.pic_h;
/* Use the frame rate from the file only if none was specified
* on the command-line.
*/
if (!arg_have_timebase)
{
cfg.g_timebase.num = y4m.fps_d;
cfg.g_timebase.den = y4m.fps_n;
}
arg_use_i420 = 0;
}
else
{
fprintf(stderr, "Unsupported Y4M stream.\n");
return EXIT_FAILURE;
}
}
else if (file_is_ivf(infile, &fourcc, &cfg.g_w, &cfg.g_h, detect.buf))
{
file_type = FILE_TYPE_IVF;
switch (fourcc)
{
case 0x32315659:
arg_use_i420 = 0;
break;
case 0x30323449:
arg_use_i420 = 1;
break;
default:
fprintf(stderr, "Unsupported fourcc (%08x) in IVF\n", fourcc);
return EXIT_FAILURE;
}
}
else
{
file_type = FILE_TYPE_RAW;
detect.valid = 1;
}
#define SHOW(field) fprintf(stderr, " %-28s = %d\n", #field, cfg.field)
if (verbose && pass == 0)
{
fprintf(stderr, "Codec: %s\n", vpx_codec_iface_name(codec->iface));
fprintf(stderr, "Source file: %s Format: %s\n", in_fn,
arg_use_i420 ? "I420" : "YV12");
fprintf(stderr, "Destination file: %s\n", out_fn);
fprintf(stderr, "Encoder parameters:\n");
SHOW(g_usage);
SHOW(g_threads);
SHOW(g_profile);
SHOW(g_w);
SHOW(g_h);
SHOW(g_timebase.num);
SHOW(g_timebase.den);
SHOW(g_error_resilient);
SHOW(g_pass);
SHOW(g_lag_in_frames);
SHOW(rc_dropframe_thresh);
SHOW(rc_resize_allowed);
SHOW(rc_resize_up_thresh);
SHOW(rc_resize_down_thresh);
SHOW(rc_end_usage);
SHOW(rc_target_bitrate);
SHOW(rc_min_quantizer);
SHOW(rc_max_quantizer);
SHOW(rc_undershoot_pct);
SHOW(rc_overshoot_pct);
SHOW(rc_buf_sz);
SHOW(rc_buf_initial_sz);
SHOW(rc_buf_optimal_sz);
SHOW(rc_2pass_vbr_bias_pct);
SHOW(rc_2pass_vbr_minsection_pct);
SHOW(rc_2pass_vbr_maxsection_pct);
SHOW(kf_mode);
SHOW(kf_min_dist);
SHOW(kf_max_dist);
}
if(pass == (one_pass_only ? one_pass_only - 1 : 0)) {
if (file_type == FILE_TYPE_Y4M)
/*The Y4M reader does its own allocation.
Just initialize this here to avoid problems if we never read any
frames.*/
memset(&raw, 0, sizeof(raw));
else
vpx_img_alloc(&raw, arg_use_i420 ? VPX_IMG_FMT_I420 : VPX_IMG_FMT_YV12,
cfg.g_w, cfg.g_h, 1);
// This was added so that ivfenc will create monotically increasing
// timestamps. Since we create new timestamps for alt-reference frames
// we need to make room in the series of timestamps. Since there can
// only be 1 alt-ref frame ( current bitstream) multiplying by 2
// gives us enough room.
cfg.g_timebase.den *= 2;
}
outfile = strcmp(out_fn, "-") ? fopen(out_fn, "wb") : stdout;
if (!outfile)
{
fprintf(stderr, "Failed to open output file\n");
return EXIT_FAILURE;
}
if (stats_fn)
{
if (!stats_open_file(&stats, stats_fn, pass))
{
fprintf(stderr, "Failed to open statistics store\n");
return EXIT_FAILURE;
}
}
else
{
if (!stats_open_mem(&stats, pass))
{
fprintf(stderr, "Failed to open statistics store\n");
return EXIT_FAILURE;
}
}
cfg.g_pass = arg_passes == 2
? pass ? VPX_RC_LAST_PASS : VPX_RC_FIRST_PASS
: VPX_RC_ONE_PASS;
#if VPX_ENCODER_ABI_VERSION > (1 + VPX_CODEC_ABI_VERSION)
if (pass)
{
cfg.rc_twopass_stats_in = stats_get(&stats);
}
#endif
write_ivf_file_header(outfile, &cfg, codec->fourcc, 0);
/* Construct Encoder Context */
vpx_codec_enc_init(&encoder, codec->iface, &cfg,
show_psnr ? VPX_CODEC_USE_PSNR : 0);
ctx_exit_on_error(&encoder, "Failed to initialize encoder");
/* Note that we bypass the vpx_codec_control wrapper macro because
* we're being clever to store the control IDs in an array. Real
* applications will want to make use of the enumerations directly
*/
for (i = 0; i < arg_ctrl_cnt; i++)
{
if (vpx_codec_control_(&encoder, arg_ctrls[i][0], arg_ctrls[i][1]))
fprintf(stderr, "Error: Tried to set control %d = %d\n",
arg_ctrls[i][0], arg_ctrls[i][1]);
ctx_exit_on_error(&encoder, "Failed to control codec");
}
frame_avail = 1;
got_data = 0;
while (frame_avail || got_data)
{
vpx_codec_iter_t iter = NULL;
const vpx_codec_cx_pkt_t *pkt;
struct vpx_usec_timer timer;
if (!arg_limit || frames_in < arg_limit)
{
frame_avail = read_frame(infile, &raw, file_type, &y4m,
&detect);
if (frame_avail)
frames_in++;
fprintf(stderr,
"\rPass %d/%d frame %4d/%-4d %7ldB \033[K", pass + 1,
arg_passes, frames_in, frames_out, nbytes);
}
else
frame_avail = 0;
vpx_usec_timer_start(&timer);
// since we halved our timebase we need to double the timestamps
// and duration we pass in.
vpx_codec_encode(&encoder, frame_avail ? &raw : NULL, (frames_in - 1) * 2,
2, 0, arg_deadline);
vpx_usec_timer_mark(&timer);
cx_time += vpx_usec_timer_elapsed(&timer);
ctx_exit_on_error(&encoder, "Failed to encode frame");
got_data = 0;
while ((pkt = vpx_codec_get_cx_data(&encoder, &iter)))
{
got_data = 1;
switch (pkt->kind)
{
case VPX_CODEC_CX_FRAME_PKT:
frames_out++;
fprintf(stderr, " %6luF",
(unsigned long)pkt->data.frame.sz);
write_ivf_frame_header(outfile, pkt);
fwrite(pkt->data.frame.buf, 1, pkt->data.frame.sz, outfile);
nbytes += pkt->data.raw.sz;
break;
case VPX_CODEC_STATS_PKT:
frames_out++;
fprintf(stderr, " %6luS",
(unsigned long)pkt->data.twopass_stats.sz);
stats_write(&stats,
pkt->data.twopass_stats.buf,
pkt->data.twopass_stats.sz);
nbytes += pkt->data.raw.sz;
break;
case VPX_CODEC_PSNR_PKT:
if (show_psnr)
{
int i;
for (i = 0; i < 4; i++)
fprintf(stderr, "%.3lf ", pkt->data.psnr.psnr[i]);
}
break;
default:
break;
}
}
fflush(stdout);
}
/* this bitrate calc is simplified and relies on the fact that this
* application uses 1/timebase for framerate.
*/
fprintf(stderr,
"\rPass %d/%d frame %4d/%-4d %7ldB %7ldb/f %7"PRId64"b/s"
" %7lu %s (%.2f fps)\033[K", pass + 1,
arg_passes, frames_in, frames_out, nbytes, nbytes * 8 / frames_in,
nbytes * 8 *(int64_t)cfg.g_timebase.den/2/ cfg.g_timebase.num / frames_in,
cx_time > 9999999 ? cx_time / 1000 : cx_time,
cx_time > 9999999 ? "ms" : "us",
(float)frames_in * 1000000.0 / (float)cx_time);
vpx_codec_destroy(&encoder);
fclose(infile);
if (!fseek(outfile, 0, SEEK_SET))
write_ivf_file_header(outfile, &cfg, codec->fourcc, frames_out);
fclose(outfile);
stats_close(&stats);
fprintf(stderr, "\n");
if (one_pass_only)
break;
}
vpx_img_free(&raw);
free(argv);
return EXIT_SUCCESS;
}
int main(int argc, char **argv) {
FILE *infile, *outfile[MAX_LAYERS];
vpx_codec_ctx_t codec;
vpx_codec_enc_cfg_t cfg;
int frame_cnt = 0;
vpx_image_t raw;
int frame_avail = 1;
int got_data = 0;
int i;
int frames_in_layer[MAX_LAYERS] = {0};
clock_t before;
clock_t after;
int pts = 0; /* PTS starts at 0 */
int frame_duration = 1; /* 1 timebase tick per frame */
parse_command_line(argc, argv, &cfg);
// Allocate image buffer
if (!vpx_img_alloc(&raw, VPX_IMG_FMT_I420, cfg.g_w, cfg.g_h, 32))
die("Failed to allocate image", cfg.g_w, cfg.g_h);
set_default_configuration(&cfg);
/* Open input file */
if (!(infile = fopen(input_filename, "rb")))
die("Failed to open %s for reading", argv[1]);
/* Open output file */
for (i = 0; i < number_spatial_layers; i++) {
char file_name[512];
snprintf(file_name, sizeof(file_name), "%s_%d.ivf", output_filename, i);
if (!(outfile[i] = fopen(file_name, "wb")))
die("Failed to open %s for writing", file_name);
write_ivf_file_header(outfile[i], &cfg, 0);
}
initialize_codec(&codec, &cfg);
// skip initial frames
for (i = 0; i < number_frames_to_skip; i++) {
read_frame(infile, &raw);
}
before = clock();
// Encoding frames
while ((frame_avail || got_data) &&
frame_cnt <= number_frames_to_code * number_spatial_layers) {
int flags = 0;
vpx_codec_iter_t iter = NULL;
const vpx_codec_cx_pkt_t *pkt;
int layer = calculate_layer(frame_cnt, number_spatial_layers);
int is_I_frame_in_layer =
(((frame_cnt - 1) / number_spatial_layers % key_period) == 0);
int is_dummy = (frame_cnt == 0);
if (is_dummy) { // Dummy frame
flags = VPX_EFLAG_FORCE_KF;
frame_avail = read_dummy_frame(&raw);
} else { // Regular frame
// Read a new frame only at the base layer
if (layer == 0) frame_avail = read_frame(infile, &raw);
switch_to_layer(layer, cfg.g_w, cfg.g_h, &codec);
flags = get_flag(is_I_frame_in_layer, layer, INTER_LAYER_PREDICTION_I);
}
// Actual Encoding
if (vpx_codec_encode(&codec, frame_avail ? &raw : NULL, pts, 1, flags,
VPX_DL_REALTIME))
die_codec(&codec, "Failed to encode frame");
got_data = 0;
// Process data / Get PSNR statistics
while ((pkt = vpx_codec_get_cx_data(&codec, &iter))) {
got_data = 1;
switch (pkt->kind) {
case VPX_CODEC_CX_FRAME_PKT:
for (i = layer; i < number_spatial_layers; i++) {
write_ivf_frame_header(outfile[i], pkt);
(void)fwrite(pkt->data.frame.buf, 1, pkt->data.frame.sz,
outfile[i]);
frames_in_layer[i]++;
}
break;
case VPX_CODEC_PSNR_PKT:
if (frame_cnt != 0)
printf(
"Processed Frame %d, layer %d, PSNR(Total/Y/U/V): "
"%2.3f %2.3f %2.3f %2.3f \n",
(frame_cnt - 1) / number_spatial_layers + 1, layer,
pkt->data.psnr.psnr[0], pkt->data.psnr.psnr[1],
pkt->data.psnr.psnr[2], pkt->data.psnr.psnr[3]);
break;
default:
break;
}
}
frame_cnt++;
// TODO(ivan): Modify ts later if(!layer)
pts += frame_duration;
}
// end while
after = clock();
printf("Processed %d frames in different resolutions in %ld ms.\n",
frame_cnt - 1, (int)(after - before) / (CLOCKS_PER_SEC / 1000));
fclose(infile);
if (vpx_codec_destroy(&codec)) die_codec(&codec, "Failed to destroy codec");
/* Try to rewrite the output file headers with the actual frame count */
for (i = 0; i < number_spatial_layers; i++) {
if (!fseek(outfile[i], 0, SEEK_SET)) {
write_ivf_file_header(outfile[i], &cfg, frames_in_layer[i]);
}
fclose(outfile[i]);
}
return EXIT_SUCCESS;
}