You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

386 lines
14KB

  1. /*
  2. * Copyright (c) 2011 Stefano Sabatini
  3. * This file is part of FFmpeg.
  4. *
  5. * FFmpeg is free software; you can redistribute it and/or
  6. * modify it under the terms of the GNU Lesser General Public
  7. * License as published by the Free Software Foundation; either
  8. * version 2.1 of the License, or (at your option) any later version.
  9. *
  10. * FFmpeg is distributed in the hope that it will be useful,
  11. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  13. * Lesser General Public License for more details.
  14. *
  15. * You should have received a copy of the GNU Lesser General Public
  16. * License along with FFmpeg; if not, write to the Free Software
  17. * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  18. */
  19. /**
  20. * @file
  21. * filter for showing textual video frame information
  22. */
  23. #include <inttypes.h>
  24. #include "libavutil/adler32.h"
  25. #include "libavutil/display.h"
  26. #include "libavutil/imgutils.h"
  27. #include "libavutil/internal.h"
  28. #include "libavutil/opt.h"
  29. #include "libavutil/pixdesc.h"
  30. #include "libavutil/spherical.h"
  31. #include "libavutil/stereo3d.h"
  32. #include "libavutil/timestamp.h"
  33. #include "libavutil/timecode.h"
  34. #include "libavutil/mastering_display_metadata.h"
  35. #include "avfilter.h"
  36. #include "internal.h"
  37. #include "video.h"
  38. typedef struct ShowInfoContext {
  39. const AVClass *class;
  40. int calculate_checksums;
  41. } ShowInfoContext;
  42. #define OFFSET(x) offsetof(ShowInfoContext, x)
  43. #define VF AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
  44. static const AVOption showinfo_options[] = {
  45. { "checksum", "calculate checksums", OFFSET(calculate_checksums), AV_OPT_TYPE_BOOL, {.i64=1}, 0, 1, VF },
  46. { NULL }
  47. };
  48. AVFILTER_DEFINE_CLASS(showinfo);
  49. static void dump_spherical(AVFilterContext *ctx, AVFrame *frame, AVFrameSideData *sd)
  50. {
  51. AVSphericalMapping *spherical = (AVSphericalMapping *)sd->data;
  52. double yaw, pitch, roll;
  53. av_log(ctx, AV_LOG_INFO, "spherical information: ");
  54. if (sd->size < sizeof(*spherical)) {
  55. av_log(ctx, AV_LOG_ERROR, "invalid data");
  56. return;
  57. }
  58. if (spherical->projection == AV_SPHERICAL_EQUIRECTANGULAR)
  59. av_log(ctx, AV_LOG_INFO, "equirectangular ");
  60. else if (spherical->projection == AV_SPHERICAL_CUBEMAP)
  61. av_log(ctx, AV_LOG_INFO, "cubemap ");
  62. else if (spherical->projection == AV_SPHERICAL_EQUIRECTANGULAR_TILE)
  63. av_log(ctx, AV_LOG_INFO, "tiled equirectangular ");
  64. else {
  65. av_log(ctx, AV_LOG_WARNING, "unknown");
  66. return;
  67. }
  68. yaw = ((double)spherical->yaw) / (1 << 16);
  69. pitch = ((double)spherical->pitch) / (1 << 16);
  70. roll = ((double)spherical->roll) / (1 << 16);
  71. av_log(ctx, AV_LOG_INFO, "(%f/%f/%f) ", yaw, pitch, roll);
  72. if (spherical->projection == AV_SPHERICAL_EQUIRECTANGULAR_TILE) {
  73. size_t l, t, r, b;
  74. av_spherical_tile_bounds(spherical, frame->width, frame->height,
  75. &l, &t, &r, &b);
  76. av_log(ctx, AV_LOG_INFO,
  77. "[%"SIZE_SPECIFIER", %"SIZE_SPECIFIER", %"SIZE_SPECIFIER", %"SIZE_SPECIFIER"] ",
  78. l, t, r, b);
  79. } else if (spherical->projection == AV_SPHERICAL_CUBEMAP) {
  80. av_log(ctx, AV_LOG_INFO, "[pad %"PRIu32"] ", spherical->padding);
  81. }
  82. }
  83. static void dump_stereo3d(AVFilterContext *ctx, AVFrameSideData *sd)
  84. {
  85. AVStereo3D *stereo;
  86. av_log(ctx, AV_LOG_INFO, "stereoscopic information: ");
  87. if (sd->size < sizeof(*stereo)) {
  88. av_log(ctx, AV_LOG_ERROR, "invalid data");
  89. return;
  90. }
  91. stereo = (AVStereo3D *)sd->data;
  92. av_log(ctx, AV_LOG_INFO, "type - %s", av_stereo3d_type_name(stereo->type));
  93. if (stereo->flags & AV_STEREO3D_FLAG_INVERT)
  94. av_log(ctx, AV_LOG_INFO, " (inverted)");
  95. }
  96. static void dump_roi(AVFilterContext *ctx, AVFrameSideData *sd)
  97. {
  98. int nb_rois;
  99. const AVRegionOfInterest *roi;
  100. uint32_t roi_size;
  101. roi = (const AVRegionOfInterest *)sd->data;
  102. roi_size = roi->self_size;
  103. if (!roi_size || sd->size % roi_size != 0) {
  104. av_log(ctx, AV_LOG_ERROR, "Invalid AVRegionOfInterest.self_size.");
  105. return;
  106. }
  107. nb_rois = sd->size / roi_size;
  108. av_log(ctx, AV_LOG_INFO, "Regions Of Interest(RoI) information: ");
  109. for (int i = 0; i < nb_rois; i++) {
  110. roi = (const AVRegionOfInterest *)(sd->data + roi_size * i);
  111. av_log(ctx, AV_LOG_INFO, "index: %d, region: (%d, %d)/(%d, %d), qp offset: %d/%d.\n",
  112. i, roi->left, roi->top, roi->right, roi->bottom, roi->qoffset.num, roi->qoffset.den);
  113. }
  114. }
  115. static void dump_mastering_display(AVFilterContext *ctx, AVFrameSideData *sd)
  116. {
  117. AVMasteringDisplayMetadata *mastering_display;
  118. av_log(ctx, AV_LOG_INFO, "mastering display: ");
  119. if (sd->size < sizeof(*mastering_display)) {
  120. av_log(ctx, AV_LOG_ERROR, "invalid data");
  121. return;
  122. }
  123. mastering_display = (AVMasteringDisplayMetadata *)sd->data;
  124. av_log(ctx, AV_LOG_INFO, "has_primaries:%d has_luminance:%d "
  125. "r(%5.4f,%5.4f) g(%5.4f,%5.4f) b(%5.4f %5.4f) wp(%5.4f, %5.4f) "
  126. "min_luminance=%f, max_luminance=%f",
  127. mastering_display->has_primaries, mastering_display->has_luminance,
  128. av_q2d(mastering_display->display_primaries[0][0]),
  129. av_q2d(mastering_display->display_primaries[0][1]),
  130. av_q2d(mastering_display->display_primaries[1][0]),
  131. av_q2d(mastering_display->display_primaries[1][1]),
  132. av_q2d(mastering_display->display_primaries[2][0]),
  133. av_q2d(mastering_display->display_primaries[2][1]),
  134. av_q2d(mastering_display->white_point[0]), av_q2d(mastering_display->white_point[1]),
  135. av_q2d(mastering_display->min_luminance), av_q2d(mastering_display->max_luminance));
  136. }
  137. static void dump_content_light_metadata(AVFilterContext *ctx, AVFrameSideData *sd)
  138. {
  139. AVContentLightMetadata* metadata = (AVContentLightMetadata*)sd->data;
  140. av_log(ctx, AV_LOG_INFO, "Content Light Level information: "
  141. "MaxCLL=%d, MaxFALL=%d",
  142. metadata->MaxCLL, metadata->MaxFALL);
  143. }
  144. static void dump_color_property(AVFilterContext *ctx, AVFrame *frame)
  145. {
  146. const char *color_range_str = av_color_range_name(frame->color_range);
  147. const char *colorspace_str = av_color_space_name(frame->colorspace);
  148. const char *color_primaries_str = av_color_primaries_name(frame->color_primaries);
  149. const char *color_trc_str = av_color_transfer_name(frame->color_trc);
  150. if (!color_range_str || frame->color_range == AVCOL_RANGE_UNSPECIFIED) {
  151. av_log(ctx, AV_LOG_INFO, "color_range:unknown");
  152. } else {
  153. av_log(ctx, AV_LOG_INFO, "color_range:%s", color_range_str);
  154. }
  155. if (!colorspace_str || frame->colorspace == AVCOL_SPC_UNSPECIFIED) {
  156. av_log(ctx, AV_LOG_INFO, " color_space:unknown");
  157. } else {
  158. av_log(ctx, AV_LOG_INFO, " color_space:%s", colorspace_str);
  159. }
  160. if (!color_primaries_str || frame->color_primaries == AVCOL_PRI_UNSPECIFIED) {
  161. av_log(ctx, AV_LOG_INFO, " color_primaries:unknown");
  162. } else {
  163. av_log(ctx, AV_LOG_INFO, " color_primaries:%s", color_primaries_str);
  164. }
  165. if (!color_trc_str || frame->color_trc == AVCOL_TRC_UNSPECIFIED) {
  166. av_log(ctx, AV_LOG_INFO, " color_trc:unknown");
  167. } else {
  168. av_log(ctx, AV_LOG_INFO, " color_trc:%s", color_trc_str);
  169. }
  170. av_log(ctx, AV_LOG_INFO, "\n");
  171. }
  172. static void update_sample_stats(const uint8_t *src, int len, int64_t *sum, int64_t *sum2)
  173. {
  174. int i;
  175. for (i = 0; i < len; i++) {
  176. *sum += src[i];
  177. *sum2 += src[i] * src[i];
  178. }
  179. }
  180. static int filter_frame(AVFilterLink *inlink, AVFrame *frame)
  181. {
  182. AVFilterContext *ctx = inlink->dst;
  183. ShowInfoContext *s = ctx->priv;
  184. const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format);
  185. uint32_t plane_checksum[4] = {0}, checksum = 0;
  186. int64_t sum[4] = {0}, sum2[4] = {0};
  187. int32_t pixelcount[4] = {0};
  188. int i, plane, vsub = desc->log2_chroma_h;
  189. for (plane = 0; plane < 4 && s->calculate_checksums && frame->data[plane] && frame->linesize[plane]; plane++) {
  190. uint8_t *data = frame->data[plane];
  191. int h = plane == 1 || plane == 2 ? AV_CEIL_RSHIFT(inlink->h, vsub) : inlink->h;
  192. int linesize = av_image_get_linesize(frame->format, frame->width, plane);
  193. if (linesize < 0)
  194. return linesize;
  195. for (i = 0; i < h; i++) {
  196. plane_checksum[plane] = av_adler32_update(plane_checksum[plane], data, linesize);
  197. checksum = av_adler32_update(checksum, data, linesize);
  198. update_sample_stats(data, linesize, sum+plane, sum2+plane);
  199. pixelcount[plane] += linesize;
  200. data += frame->linesize[plane];
  201. }
  202. }
  203. av_log(ctx, AV_LOG_INFO,
  204. "n:%4"PRId64" pts:%7s pts_time:%-7s pos:%9"PRId64" "
  205. "fmt:%s sar:%d/%d s:%dx%d i:%c iskey:%d type:%c ",
  206. inlink->frame_count_out,
  207. av_ts2str(frame->pts), av_ts2timestr(frame->pts, &inlink->time_base), frame->pkt_pos,
  208. desc->name,
  209. frame->sample_aspect_ratio.num, frame->sample_aspect_ratio.den,
  210. frame->width, frame->height,
  211. !frame->interlaced_frame ? 'P' : /* Progressive */
  212. frame->top_field_first ? 'T' : 'B', /* Top / Bottom */
  213. frame->key_frame,
  214. av_get_picture_type_char(frame->pict_type));
  215. if (s->calculate_checksums) {
  216. av_log(ctx, AV_LOG_INFO,
  217. "checksum:%08"PRIX32" plane_checksum:[%08"PRIX32,
  218. checksum, plane_checksum[0]);
  219. for (plane = 1; plane < 4 && frame->data[plane] && frame->linesize[plane]; plane++)
  220. av_log(ctx, AV_LOG_INFO, " %08"PRIX32, plane_checksum[plane]);
  221. av_log(ctx, AV_LOG_INFO, "] mean:[");
  222. for (plane = 0; plane < 4 && frame->data[plane] && frame->linesize[plane]; plane++)
  223. av_log(ctx, AV_LOG_INFO, "%"PRId64" ", (sum[plane] + pixelcount[plane]/2) / pixelcount[plane]);
  224. av_log(ctx, AV_LOG_INFO, "\b] stdev:[");
  225. for (plane = 0; plane < 4 && frame->data[plane] && frame->linesize[plane]; plane++)
  226. av_log(ctx, AV_LOG_INFO, "%3.1f ",
  227. sqrt((sum2[plane] - sum[plane]*(double)sum[plane]/pixelcount[plane])/pixelcount[plane]));
  228. av_log(ctx, AV_LOG_INFO, "\b]");
  229. }
  230. av_log(ctx, AV_LOG_INFO, "\n");
  231. for (i = 0; i < frame->nb_side_data; i++) {
  232. AVFrameSideData *sd = frame->side_data[i];
  233. av_log(ctx, AV_LOG_INFO, " side data - ");
  234. switch (sd->type) {
  235. case AV_FRAME_DATA_PANSCAN:
  236. av_log(ctx, AV_LOG_INFO, "pan/scan");
  237. break;
  238. case AV_FRAME_DATA_A53_CC:
  239. av_log(ctx, AV_LOG_INFO, "A/53 closed captions (%d bytes)", sd->size);
  240. break;
  241. case AV_FRAME_DATA_SPHERICAL:
  242. dump_spherical(ctx, frame, sd);
  243. break;
  244. case AV_FRAME_DATA_STEREO3D:
  245. dump_stereo3d(ctx, sd);
  246. break;
  247. case AV_FRAME_DATA_S12M_TIMECODE: {
  248. uint32_t *tc = (uint32_t*)sd->data;
  249. for (int j = 1; j <= tc[0]; j++) {
  250. char tcbuf[AV_TIMECODE_STR_SIZE];
  251. av_timecode_make_smpte_tc_string(tcbuf, tc[j], 0);
  252. av_log(ctx, AV_LOG_INFO, "timecode - %s%s", tcbuf, j != tc[0] ? ", " : "");
  253. }
  254. break;
  255. }
  256. case AV_FRAME_DATA_DISPLAYMATRIX:
  257. av_log(ctx, AV_LOG_INFO, "displaymatrix: rotation of %.2f degrees",
  258. av_display_rotation_get((int32_t *)sd->data));
  259. break;
  260. case AV_FRAME_DATA_AFD:
  261. av_log(ctx, AV_LOG_INFO, "afd: value of %"PRIu8, sd->data[0]);
  262. break;
  263. case AV_FRAME_DATA_REGIONS_OF_INTEREST:
  264. dump_roi(ctx, sd);
  265. break;
  266. case AV_FRAME_DATA_MASTERING_DISPLAY_METADATA:
  267. dump_mastering_display(ctx, sd);
  268. break;
  269. case AV_FRAME_DATA_CONTENT_LIGHT_LEVEL:
  270. dump_content_light_metadata(ctx, sd);
  271. break;
  272. case AV_FRAME_DATA_GOP_TIMECODE: {
  273. char tcbuf[AV_TIMECODE_STR_SIZE];
  274. av_timecode_make_mpeg_tc_string(tcbuf, *(int64_t *)(sd->data));
  275. av_log(ctx, AV_LOG_INFO, "GOP timecode - %s", tcbuf);
  276. break;
  277. }
  278. default:
  279. av_log(ctx, AV_LOG_WARNING, "unknown side data type %d (%d bytes)",
  280. sd->type, sd->size);
  281. break;
  282. }
  283. av_log(ctx, AV_LOG_INFO, "\n");
  284. }
  285. dump_color_property(ctx, frame);
  286. return ff_filter_frame(inlink->dst->outputs[0], frame);
  287. }
  288. static int config_props(AVFilterContext *ctx, AVFilterLink *link, int is_out)
  289. {
  290. av_log(ctx, AV_LOG_INFO, "config %s time_base: %d/%d, frame_rate: %d/%d\n",
  291. is_out ? "out" : "in",
  292. link->time_base.num, link->time_base.den,
  293. link->frame_rate.num, link->frame_rate.den);
  294. return 0;
  295. }
  296. static int config_props_in(AVFilterLink *link)
  297. {
  298. AVFilterContext *ctx = link->dst;
  299. return config_props(ctx, link, 0);
  300. }
  301. static int config_props_out(AVFilterLink *link)
  302. {
  303. AVFilterContext *ctx = link->src;
  304. return config_props(ctx, link, 1);
  305. }
  306. static const AVFilterPad avfilter_vf_showinfo_inputs[] = {
  307. {
  308. .name = "default",
  309. .type = AVMEDIA_TYPE_VIDEO,
  310. .filter_frame = filter_frame,
  311. .config_props = config_props_in,
  312. },
  313. { NULL }
  314. };
  315. static const AVFilterPad avfilter_vf_showinfo_outputs[] = {
  316. {
  317. .name = "default",
  318. .type = AVMEDIA_TYPE_VIDEO,
  319. .config_props = config_props_out,
  320. },
  321. { NULL }
  322. };
  323. AVFilter ff_vf_showinfo = {
  324. .name = "showinfo",
  325. .description = NULL_IF_CONFIG_SMALL("Show textual information for each video frame."),
  326. .inputs = avfilter_vf_showinfo_inputs,
  327. .outputs = avfilter_vf_showinfo_outputs,
  328. .priv_size = sizeof(ShowInfoContext),
  329. .priv_class = &showinfo_class,
  330. };