Bad cost tables used in ARNR filtering.

The use of incorrect mv costing tables in the ARNR sub-pel
filtering code led to corruption of the altref buffer in some cases,
particularly at low data rates.

The average gain from this fix is about 0.3% but there are a few
extreme cases where nasty and visible artifacts manifested and
for these few data points the improvement is > 10%.

PGW and AWG

Change-Id: I95cc02b196a433e71d0d2bd2b933fe68ed31e796

diff --git a/vp8/encoder/temporal_filter.c b/vp8/encoder/temporal_filter.c
index fd5dd7edeb2be96e555b0bc90b533febca28f824..abc50270c045134282173755feaadd90aeab45c1 100644 (file)
--- a/vp8/encoder/temporal_filter.c
+++ b/vp8/encoder/temporal_filter.c
@@ -292,7 +292,7 @@ static int find_matching_mb
         bestsme = cpi->find_fractional_mv_step(x, b, d,
                     &d->bmi.mv.as_mv, &best_ref_mv1,
                     x->errorperbit, &cpi->fn_ptr[BLOCK_16X16],
-                    cpi->mb.mvcost);
+                    mvcost);
     }
 #endif