shithub: libvpx

Info  •  Files  •  Log  •  Branches

Use precise context to estimate coeff rate cost

Use the precise context to estimate the zero token cost in trellis
optimization process. This improves the speed 0 coding performance
by 0.15% for lowres and 0.1% for midres. It improves the speed 1
coding performance by 0.2% for midres and hdres.

Change-Id: I59c7c08702fc79dc4f8534b64ca594da909e2c91

--- a/vp9/encoder/vp9_encodemb.c

+++ b/vp9/encoder/vp9_encodemb.c

@@ -314,17 +314,18 @@

        *  add a new trellis node, but we do need to update the costs.

        */

       band = band_translate[i + 1];

+      pt = get_coef_context(nb, token_cache, i + 1);

       t0 = tokens[next][0].token;

       t1 = tokens[next][1].token;

       /* Update the cost of each path if we're past the EOB token. */

       if (t0 != EOB_TOKEN) {

         tokens[next][0].rate +=

-            mb->token_costs[tx_size][type][ref][band][1][0][t0];

+            mb->token_costs[tx_size][type][ref][band][1][pt][t0];

         tokens[next][0].token = ZERO_TOKEN;

       }

       if (t1 != EOB_TOKEN) {

         tokens[next][1].rate +=

-            mb->token_costs[tx_size][type][ref][band][1][0][t1];

+            mb->token_costs[tx_size][type][ref][band][1][pt][t1];

         tokens[next][1].token = ZERO_TOKEN;

       }

       best_index[i][0] = best_index[i][1] = 0;