shithub: libvpx

Info  •  Files  •  Log  •  Branches

change the threshold of DC check for encode breakout

Previously, the DC check is to make sure there is no code-able
DC shift for quantizer Q0, which has been verified rather
conservative. This commit changes the criteria to have two
components, DC and AC, to address the conservativeness. First,
it checks if all AC energy is enough to contribute a single
non-zero quantized AC coefficient. Second, for DC, the decision
to skip further considers two possible scenarios: 1. There is
no code-able 2nd order DC coefficient at all; 2 The residue is
relatively flat, but the uniform DC change is very small, i.e.
less than 1/2 gray level per pixel.

Comparing to previous criteria, the new criteria is about 10%
to 15% faster in encoding time with a very small quality loss.
(threshold ~1000 and quality range 33db-45db)

It should be noted that this commit enables "automatic" static
threshold for encodebreakout if a non-zero small value is passed
in to encoder.

Change-Id: I0f77719a1ac2c2dfddbd950d84920df374515ce3

--- a/vp8/encoder/rdopt.c

+++ b/vp8/encoder/rdopt.c

@@ -2267,22 +2267,28 @@

             else if (x->encode_breakout)

             {

                 int sum, sse;

+                int threshold = (xd->block[0].dequant[1]

+                            * xd->block[0].dequant[1] >>4);

+                if(threshold < x->encode_breakout)

+                    threshold = x->encode_breakout;

+

                 VARIANCE_INVOKE(&cpi->rtcd.variance, get16x16var)

                     (x->src.y_buffer, x->src.y_stride,

                      x->e_mbd.predictor, 16, (unsigned int *)(&sse), &sum);

-                if (sse < x->encode_breakout)

+                if (sse < threshold)

                 {

                     // Check u and v to make sure skip is ok

                     int sse2 = 0;

-

-                    // add dc check

-                    if (abs(sum) < (cpi->common.Y2dequant[0][0] << 2))

+                    /* If theres is no codeable 2nd order dc

+                       or a very small uniform pixel change change */

+                    if (abs(sum) < (xd->block[24].dequant[0]<<2)||

+                        ((sum * sum>>8) > sse && abs(sum) <128))

                     {

                         sse2 = VP8_UVSSE(x, IF_RTCD(&cpi->rtcd.variance));

-                        if (sse2 * 2 < x->encode_breakout)

+                        if (sse2 * 2 < threshold)

                         {

                             x->skip = 1;

                             distortion2 = sse + sse2;

@@ -2428,6 +2434,7 @@

         if (x->skip)

             break;

+

     }

     // Reduce the activation RD thresholds for the best choice mode

-- 

⑨