Intel® Integrated Performance Primitives


I have used the FIR and FIRLMS functions, however it appears that when I use FIRLMSMR functions I cannot get an output other than zero from the funtion. Have these algorithms been completed or are they not yet finished?

I use the InitAlloc function to load in filter taps and a delay line, and I set the appropriate paramters, then I call then main FIRLMSMR function to compute an output and it always appears to be zero.

Anyone else have luck with these functions?


DCT in ipp_sample MPEG2 for Linux


I want to change the DCT coefficients from the encoder.
When I tray do
ippiQuantIntra_MPEG2_16s_C1I( block, pContext->Video[pContext->cvindex].QScale, 0, count);
for (c=0;c<65;c++) block[c]=((block[c]>>1)<<1)|1;

it's ok, but when I take off |1

for (c=0;c<65;c++) block[c]=((block[c]>>1)<<1);

I have the message
qtplayer: video/seq.c:1818: Macroblock_420_I_frame: Assertion `ret == 0' failed.

What's happening? Am I in wrong position form encoder?

LUT Function Question...

Hello everyone!

We're currently using IPP 4.1 on one of our projects, and it's been an overall plus in terms of overall performance and efficiency!

My question is in regards to the set of ippiLUT_X functions in the image processing library. What I'd like to do is apply an intensity transformation where each pixel of the image has it's own set of levels and values. I'm wondering if this is even possible?

Extracting JPEG thumbnails


I have an application that needs to parse thousands of JPEGs as quickly as possible. From these JPEGs all I need is the EXIF thumbnail (a 160x120 image).

I was looking at the IPP JPEG library and samples, and it all seems very complex with lots of initialization that are relevant for decompressing the whole image.

Could anyone give me a small example as to how I can extract the thumbnail into my own 160x120 8u3C image buffer?

Note: the thumbnail itself is also compressed in JPEG so I need to decompress that into my raw pixel buffer.

Why isn't it faster?


I have a c++ Vector4d class and was hoping to improve its performance using ipp. I had 4 doubles with the values and exchanged them for a vector (which I allocate with ippsMalloc).

Now, I only perform regular operations (+, -, dot product) and I tested the class against its old implementation, using the dot product for that (its the operation I use the most):

Using g++ without optimization flags, I got slightly better results with ipp. With compiler optimization, though, the code without ipp was a lot faster (!!!).

require some help about ipp video coding function


I have called the function

ippiFilterDeblocking_HorEdge_MPEG4_8u_I(DecDec->image_ref->y, 176,8,2,6);

with the parameters as specified.
It is a deblocking filter but it did not removed the blocks from video.
I have followed the standard procedure function is called but results are there

plz tell me whether i should call some suportting or intialization functions before calling the function.
or i have correctly used the function.
Is the function is coded correctly or there are some bugs in its functionallity in ipp.

JPEG encoder question

I've tried both the new IPP-based IJL (ijl2.0) and the CJPEGEncoder class.
Interestingly I've found that the jpeg file size is much larger when you do a save in the jpegviewer based on CJPEGEncoder class. For example, I loaded a 169,514 bytes jpeg file. Once I clicked on the disk icon in the toolbar, the same filebecame 3,198,768 bytes. Something is not right. The jpgview program based on JIL-IPP does not have the same problem.

订阅 Intel® Integrated Performance Primitives