From: Jingning Han <jingning@google.com>
Date: Wed, 10 Oct 2018 21:52:30 +0000 (-0700)
Subject: Call tpl model build at the beginning of a GOP
X-Git-Tag: v1.8.0~248^2
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=d0f7f002b13081ed016033741a021987f70ac568;p=libvpx

Call tpl model build at the beginning of a GOP

The gop index 0 is default as kf / gf. The effective first coding
frame controlled by the current GOP rate allocation is indexed 1.
Call the tpl model build for the current GOP once at index 1
position. This would unify the calling system for single/multi-layer
ARF GOP structure.

Change-Id: I4ce69337e04646098d5513c0aa56b4e0b4483337
---

diff --git a/vp9/encoder/vp9_encoder.c b/vp9/encoder/vp9_encoder.c
index 4a55e08a4..8e35a8bad 100644
--- a/vp9/encoder/vp9_encoder.c
+++ b/vp9/encoder/vp9_encoder.c
@@ -6373,7 +6373,7 @@ int vp9_get_compressed_data(VP9_COMP *cpi, unsigned int *frame_flags,
     for (i = 0; i < MAX_REF_FRAMES; ++i) cpi->scaled_ref_idx[i] = INVALID_IDX;
   }
 
-  if (arf_src_index && cpi->sf.enable_tpl_model) {
+  if (cpi->twopass.gf_group.index == 1 && cpi->sf.enable_tpl_model) {
     vp9_estimate_qp_gop(cpi);
     setup_tpl_stats(cpi);
   }