Convert NV12 to RGB32 using SSE2

Source: Internet
Author: User

Recently, we used Intel's media SDK to play H264 videos. We found that the frames transcoded by intel's SDK are in the NV12 pixel format and must be converted to the RGB32 format by ourselves. Although this type of code looks for a lot on the Internet, you can do it yourself.

Google's conversion formula:

C = Y-16
D = U-128
E = V-128
R = clip (round (1.164383 * C + 1.596027 * E ))
G = clip (round (1.164383 * C-(0.391762 * D)-(0.812968 * E )))
B = clip (round (1.164383 * C + 2.017232 * D ))

I studied how to use SSE2 for an hour. I tried to debug the code once, and I was in a good mood. I pasted the core code.

 

_ M128i c0 = _ mm_setzero_si128 ();
_ M128i c128 = _ mm_setpaiepi16 (128 );
_ M128i c128_32 = _ mm_setpaiepi32 (128 );
_ M128i c16 = _ mm_setpaiepi16 (16 );
_ M128i c255 = _ mm_setpaiepi16 (255 );
_ M128i c_1_1596 = _ mm_set1_epi32 (0x199012a );
_ M128i c_1_2017 = _ mm_set1_epi32 (0x204012a );
_ M128i c_0_392 = _ mm_set1_epi32 (0xff9c0000 );
_ M128i c_0000813 = _ mm_set0000epi32 (0xff30012a );
For (int y = 0; y <src. Height; y ++)
{
BYTE * dest = (BYTE *) data. Scan0 + data. Stride * y;
BYTE * srcY = src. Y + src. Pitch * y;
BYTE * srcUV = src. UV + src. Pitch * (y/2 );
For (int x = 0; x <src. Width; x + = 4)
{
// Y0Y1Y2Y30000-16
_ M128i Ymm = _ mm_sub_epi16 (_ mm_unpacklo_epi8 (_ mm_cvtsi32_si128 (* (int *) (srcY + x), c0), c16 );
// U0V0U2V20000-128
_ M128i UVmm = _ mm_sub_epi16 (_ mm_unpacklo_epi8 (_ mm_cvtsi32_si128 (* (int *) (srcUV + x), c0), c128 );
// U0U0U2U20000
_ M128i Umm = _ mm_shufflelo_epi16 (UVmm, _ MM_SHUFFLE (2, 2, 0, 0 ));
// V0V0V2V20000
_ M128i Vmm = _ mm_shufflelo_epi16 (UVmm, _ MM_SHUFFLE (3, 3, 1, 1 ));
// Y0V0Y1V0Y2V2Y3V2
_ M128i YVmm = _ mm_unpacklo_epi16 (Ymm, Vmm );
// Y0U0Y1U0Y2U2Y3U2
_ M128i YUmm = _ mm_unpacklo_epi16 (Ymm, Umm );

_ M128i Rmm = _ mm_srai_epi32 (_ mm_add_epi32 (_ mm_madd_epi16 (YVmm, c_00001596), c128_32), 8 );
_ M128i Bmm = _ mm_srai_epi32 (_ mm_add_epi32 (_ mm_madd_epi16 (YUmm, c_1_2017), c128_32), 8 );
_ M128i Gmm = _ second (_ mm_madd_epi16 (YVmm, c_1_813), _ mm_madd_epi16 (YUmm, c_0_392), c128_32), 8 );
Rmm = _ mm_slli_epi32 (_ mm_and_si128 (Rmm, _ mm_cmpgt_epi32 (Rmm, c0), 16 );
Bmm = _ mm_and_si128 (Bmm, _ mm_cmpgt_epi32 (Bmm, c0 ));
Gmm = _ mm_slli_epi32 (_ mm_min_epi16 (_ mm_and_si128 (Gmm, _ mm_cmpgt_epi32 (Gmm, c0), c255), 8 );
* (_ M128i *) dest = _ mm_or_si128 (_ mm_min_epi16 (_ mm_or_si128 (Rmm, Bmm), c255), Gmm );
Dest + = 16;
}
}

 

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.