Dual 16x16 inter prediction.

This patch introduces the concept of dual inter16x16 prediction. A
16x16 inter-predicted macroblock can use 2 references instead of 1,
where both references use the same mvmode (new, near/est, zero). In the
case of newmv, this means that two MVs are coded instead of one. The
frame can be encoded in 3 ways: all MBs single-prediction, all MBs dual
prediction, or per-MB single/dual prediction selection ("hybrid"), in
which case a single bit is coded per-MB to indicate whether the MB uses
single or dual inter prediction.

In the future, we can (maybe?) get further gains by mixing this with
Adrian's 32x32 work, per-segment dual prediction settings, or adding
support for dual splitmv/8x8mv inter prediction.

Gain (on derf-set, CQ mode) is ~2.8% (SSIM) or ~3.6% (glb PSNR). Most
gain is at medium/high bitrates, but there's minor gains at low bitrates
also. Output was confirmed to match between encoder and decoder.

Note for optimization people: this patch introduces a 2nd version of
16x16/8x8 sixtap/bilin functions, which does an avg instead of a
store. They may want to look and make sure this is implemented to
their satisfaction so we can optimize it best in the future.

Change-ID: I59dc84b07cbb3ccf073ac0f756d03d294cb19281

1 files changed, 20 insertions, 8 deletions

diff --git a/vp8/common/generic/systemdependent.c b/vp8/common/generic/systemdependent.c
index 68ed8aab0..961916314 100644
--- a/vp8/common/generic/systemdependent.c
+++ b/vp8/common/generic/systemdependent.c
@@ -84,6 +84,10 @@ void vp8_machine_specific_config(VP8_COMMON *ctx)
 #endif
     rtcd->recon.copy16x16   = vp8_copy_mem16x16_c;
     rtcd->recon.copy8x8     = vp8_copy_mem8x8_c;
+#if CONFIG_DUALPRED
+    rtcd->recon.avg16x16    = vp8_avg_mem16x16_c;
+    rtcd->recon.avg8x8      = vp8_avg_mem8x8_c;
+#endif /* CONFIG_DUALPRED */
     rtcd->recon.copy8x4     = vp8_copy_mem8x4_c;
     rtcd->recon.recon       = vp8_recon_b_c;
 #if CONFIG_I8X8
@@ -112,14 +116,22 @@ void vp8_machine_specific_config(VP8_COMMON *ctx)
 #endif
 
 
-    rtcd->subpix.sixtap16x16   = vp8_sixtap_predict16x16_c;
-    rtcd->subpix.sixtap8x8     = vp8_sixtap_predict8x8_c;
-    rtcd->subpix.sixtap8x4     = vp8_sixtap_predict8x4_c;
-    rtcd->subpix.sixtap4x4     = vp8_sixtap_predict_c;
-    rtcd->subpix.bilinear16x16 = vp8_bilinear_predict16x16_c;
-    rtcd->subpix.bilinear8x8   = vp8_bilinear_predict8x8_c;
-    rtcd->subpix.bilinear8x4   = vp8_bilinear_predict8x4_c;
-    rtcd->subpix.bilinear4x4   = vp8_bilinear_predict4x4_c;
+    rtcd->subpix.sixtap16x16       = vp8_sixtap_predict16x16_c;
+    rtcd->subpix.sixtap8x8         = vp8_sixtap_predict8x8_c;
+#if CONFIG_DUALPRED
+    rtcd->subpix.sixtap_avg16x16   = vp8_sixtap_predict_avg16x16_c;
+    rtcd->subpix.sixtap_avg8x8     = vp8_sixtap_predict_avg8x8_c;
+#endif /* CONFIG_DUALPRED */
+    rtcd->subpix.sixtap8x4         = vp8_sixtap_predict8x4_c;
+    rtcd->subpix.sixtap4x4         = vp8_sixtap_predict_c;
+    rtcd->subpix.bilinear16x16     = vp8_bilinear_predict16x16_c;
+    rtcd->subpix.bilinear8x8       = vp8_bilinear_predict8x8_c;
+#if CONFIG_DUALPRED
+    rtcd->subpix.bilinear_avg16x16 = vp8_bilinear_predict_avg16x16_c;
+    rtcd->subpix.bilinear_avg8x8   = vp8_bilinear_predict_avg8x8_c;
+#endif /* CONFIG_DUALPRED */
+    rtcd->subpix.bilinear8x4       = vp8_bilinear_predict8x4_c;
+    rtcd->subpix.bilinear4x4       = vp8_bilinear_predict4x4_c;
 
     rtcd->loopfilter.normal_mb_v = vp8_loop_filter_mbv_c;
     rtcd->loopfilter.normal_b_v  = vp8_loop_filter_bv_c;