Commit 2aa6117b authored by Jingning Han's avatar Jingning Han
Browse files

Refactor transform selection process

This commit re-arranges the transform type and size selectio
process. It removes an unnecessary rate-distortion cost computation
step. Local experiments show that this speeds up the encoding
process by 6% for both the baseline and the ext-intra experiment.

Change-Id: Iab3b86a63a1e9e55548466791ed5d29a0575c1e7
parent c5449d3e
......@@ -1681,12 +1681,6 @@ static void choose_tx_size_from_rd(VP10_COMP *cpi, MACROBLOCK *x,
if (mbmi->tx_size >= TX_32X32)
assert(mbmi->tx_type == DCT_DCT);
&r, &d, &s,
&sse, ref_best_rd, 0, bs, best_tx,
static void super_block_yrd(VP10_COMP *cpi, MACROBLOCK *x, int *rate,
......@@ -8249,6 +8243,7 @@ void vp10_rd_pick_inter_mode_sb(VP10_COMP *cpi,
if (this_mode != DC_PRED && this_mode != TM_PRED)
rate2 += intra_cost_penalty;
distortion2 = distortion_y + distortion_uv;
vp10_encode_intra_block_plane(x, bsize, 0);
if (xd->cur_buf->flags & YV12_FLAG_HIGHBITDEPTH) {
x->recon_variance =
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment