Quantcast
Channel: Intel® Integrated Performance Primitives
Viewing all articles
Browse latest Browse all 1294

ippiFilterBorder bug

$
0
0

I noticed a bug in ippiFilterBorder where ippiFilterBorder_16u_C4R and ippiFilterBorder_16u_C1R (and rarely also ippiFilterBorder_32f_C4R and ippiFilterBorder_32f_C1R, but never ippiFilterBorder_8u_C4R or ippiFilterBorder_8u_C1R) write random pixels in parts of the last row of the target image when the convolution kernel has a width of 1 and a height > 1. The result seems to depend on the border type, but does happen with all border types.

The workaround is to apply a horizontal kernel (with height 1 and width > 1) to a transposed image and then to transpose the result back.

This happens with IPP version 9.0, but I can't reproduce it with version 2018-2 or 2019-0. However, I don't see an entry for this in the IPP bug fix logs. Can you confirm that this is something that has been fixed between version 9.0 and 2018-2 ?

Regards,

Adriaan van Os

 

TCE Open Date: 

Wednesday, March 11, 2020 - 04:13

Viewing all articles
Browse latest Browse all 1294

Latest Images

Trending Articles



Latest Images

<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>