vpxenc: support scaling prior to encoding

Scales the input of the encoder using libyuv's "box filter". Each stream
may have a different width and height specified. If the width (or
height) parameter is missing (or is explicitly set to 0) then the value
will be calculated based on the specified height (or width) and the
input file's dimensions, preserving its aspect ratio. Leaving the height
unspecified behaves similarly.

Change-Id: Ic7026810b13be030826be80dc6f7fc4aaf0c35d0

diff --git a/examples.mk b/examples.mk
index f1cc42bf7b3ca6cfdf8ca9bd5430e6b8a03fd6ce..28430a611311c843b81be7c8dec8d73068b85eea 100644 (file)
--- a/examples.mk
+++ b/examples.mk
@@ -8,6 +8,12 @@
 ##  be found in the AUTHORS file in the root of the source tree.
 ##
 
+LIBYUV_SRCS +=  third_party/libyuv/include/libyuv/basic_types.h  \
+                third_party/libyuv/include/libyuv/cpu_id.h  \
+                third_party/libyuv/include/libyuv/scale.h  \
+                third_party/libyuv/source/row.h \
+                third_party/libyuv/source/scale.c  \
+                third_party/libyuv/source/cpu_id.c
 
 # List of examples to build. UTILS are files that are taken from the source
 # tree directly, and GEN_EXAMPLES are files that are created from the
@@ -36,6 +42,7 @@ vpxenc.SRCS                 += vpx_ports/vpx_timer.h
 vpxenc.SRCS                 += libmkv/EbmlIDs.h
 vpxenc.SRCS                 += libmkv/EbmlWriter.c
 vpxenc.SRCS                 += libmkv/EbmlWriter.h
+vpxenc.SRCS                 += $(LIBYUV_SRCS)
 vpxenc.GUID                  = 548DEC74-7A15-4B2B-AFC3-AA102E7C25C1
 vpxenc.DESCRIPTION           = Full featured encoder
 UTILS-$(CONFIG_VP8_ENCODER)    += vp8_scalable_patterns.c
@@ -99,13 +106,7 @@ vp8cx_set_ref.DESCRIPTION           = VP8 set encoder reference frame
 
 # C file is provided, not generated automatically.
 UTILS-$(CONFIG_MULTI_RES_ENCODING) += vp8_multi_resolution_encoder.c
-vp8_multi_resolution_encoder.SRCS  \
-                         += third_party/libyuv/include/libyuv/basic_types.h  \
-                            third_party/libyuv/include/libyuv/cpu_id.h  \
-                            third_party/libyuv/include/libyuv/scale.h  \
-                            third_party/libyuv/source/row.h \
-                            third_party/libyuv/source/scale.c  \
-                            third_party/libyuv/source/cpu_id.c
+vp8_multi_resolution_encoder.SRCS         += $(LIBYUV_SRCS)
 vp8_multi_resolution_encoder.GUID         = 04f8738e-63c8-423b-90fa-7c2703a374de
 vp8_multi_resolution_encoder.DESCRIPTION  = VP8 Multiple-resolution Encoding
 

diff --git a/vpxenc.c b/vpxenc.c
index 2f3b6356dd14b819e4cd60491fdbdf2373f47ae1..7597e3cbbf36dad658ab89dcdab908590ef0dbf3 100644 (file)
--- a/vpxenc.c
+++ b/vpxenc.c
@@ -47,6 +47,7 @@
 #include "y4minput.h"
 #include "libmkv/EbmlWriter.h"
 #include "libmkv/EbmlIDs.h"
+#include "third_party/libyuv/include/libyuv/scale.h"
 
 /* Need special handling of these functions on Windows */
 #if defined(_MSC_VER)
@@ -1642,6 +1643,7 @@ struct stream_state {
   uint64_t                  cx_time;
   size_t                    nbytes;
   stats_io_t                stats;
+  struct vpx_image         *img;
   vpx_codec_ctx_t           decoder;
   vpx_ref_frame_t           ref_enc;
   vpx_ref_frame_t           ref_dec;
@@ -2061,11 +2063,15 @@ static void validate_stream_config(struct stream_state *stream) {
 static void set_stream_dimensions(struct stream_state *stream,
                                   unsigned int w,
                                   unsigned int h) {
-  if ((stream->config.cfg.g_w && stream->config.cfg.g_w != w)
-      || (stream->config.cfg.g_h && stream->config.cfg.g_h != h))
-    fatal("Stream %d: Resizing not yet supported", stream->index);
-  stream->config.cfg.g_w = w;
-  stream->config.cfg.g_h = h;
+  if (!stream->config.cfg.g_w) {
+    if (!stream->config.cfg.g_h)
+      stream->config.cfg.g_w = w;
+    else
+      stream->config.cfg.g_w = w * stream->config.cfg.g_h / h;
+  }
+  if (!stream->config.cfg.g_h) {
+    stream->config.cfg.g_h = h * stream->config.cfg.g_w / w;
+  }
 }
 
 
@@ -2258,6 +2264,28 @@ static void encode_frame(struct stream_state  *stream,
   next_frame_start = (cfg->g_timebase.den * (int64_t)(frames_in)
                       * global->framerate.den)
                      / cfg->g_timebase.num / global->framerate.num;
+
+  /* Scale if necessary */
+  if (img && (img->d_w != cfg->g_w || img->d_h != cfg->g_h)) {
+    if (!stream->img)
+      stream->img = vpx_img_alloc(NULL, VPX_IMG_FMT_I420,
+                                  cfg->g_w, cfg->g_h, 16);
+    I420Scale(img->planes[VPX_PLANE_Y], img->stride[VPX_PLANE_Y],
+              img->planes[VPX_PLANE_U], img->stride[VPX_PLANE_U],
+              img->planes[VPX_PLANE_V], img->stride[VPX_PLANE_V],
+              img->d_w, img->d_h,
+              stream->img->planes[VPX_PLANE_Y],
+              stream->img->stride[VPX_PLANE_Y],
+              stream->img->planes[VPX_PLANE_U],
+              stream->img->stride[VPX_PLANE_U],
+              stream->img->planes[VPX_PLANE_V],
+              stream->img->stride[VPX_PLANE_V],
+              stream->img->d_w, stream->img->d_h,
+              kFilterBox);
+
+    img = stream->img;
+  }
+
   vpx_usec_timer_start(&timer);
   vpx_codec_encode(&stream->encoder, img, frame_start,
                    (unsigned long)(next_frame_start - frame_start),
@@ -2518,6 +2546,9 @@ int main(int argc, const char **argv_) {
     });
 
     /* Update stream configurations from the input file's parameters */
+    if (!input.w || !input.h)
+      fatal("Specify stream dimensions with --width (-w) "
+            " and --height (-h)");
     FOREACH_STREAM(set_stream_dimensions(stream, input.w, input.h));
     FOREACH_STREAM(validate_stream_config(stream));