Backgound Blur

Andrej Valencic andrej.valencic at zoho.com
Wed Dec 17 14:06:19 GMT 2025


Done!

On 17/12/2025 14:33, Gilles Caulier wrote:
> Hi,
>
> The algorithm used in Background Blur can be certainly improved
> especially in speed form. It's based on OpenCV API, which uses GPU if
> available to speedup processing of image data. Currently OpenCV is
> optimized for NVIDIA and Apple Silicon GPU. Here time processing is
> human compatible with HEIF photos taken with my iphone 15 pro (48Mpx).
>
> I recommend creating a new file in bugzilla about this topic in the
> Plugin-Editor-BackgroundBlur section.
>
> Best regards
>
> Gilles Caulier
>
> Le lun. 15 déc. 2025 à 18:06, Andrej Valencic
> <andrej.valencic at zoho.com> a écrit :
>> Hello!
>>
>> I found out about this new tool - Background Blur - and was very excited
>> to try it out.
>> Unfortunately in this present form this tool is not very useful as it
>> takes 10+ minutes to complete the transformation. Even changing some of
>> the parameters and checking the result is very slooooow. Is this normal
>> and are there any plans/possibilities to make this tool more responsive?
>>
>> I'm using the latest build of DK on my Windows 11 computer  that is
>> quite powerful (see bellow).
>>
>> CPU
>>
>>       AMD Ryzen 7 7735HS with Radeon Graphics
>>
>>       Base speed:    3,20 GHz
>>       Sockets:    1
>>       Cores:    8
>>       Logical processors:    16
>>       Virtualization:    Enabled
>>       L1 cache:    512 KB
>>       L2 cache:    4,0 MB
>>       L3 cache:    16,0 MB
>>
>>       Utilization    2%
>>       Speed    3,64 GHz
>>       Up time    2:07:45:53
>>       Processes    219
>>       Threads    3868
>>       Handles    142036
>>
>> Memory
>>
>>       32,0 GB
>>
>>       Speed:    4800 MT/s
>>       Slots used:    2 of 2
>>       Form factor:    SODIMM
>>       Hardware reserved:    3,3 GB
>>
>>       Available    15,1 GB
>>       Cached    12,3 GB
>>       Committed    24,5/30,6 GB
>>       Paged pool    753 MB
>>       Non-paged pool    700 MB
>>       In use (Compressed)    13,5 GB (1,8 GB)
>>
>>



More information about the Digikam-users mailing list