[digiKam-users] GSOC 2017 Project 4 : digiKam Face Management improvements added to DK6?

[NeiNei]

Dear Gilles and team,

I just read about the DK 6 beta and remembered then GSOC 2017 Project 4
digiKam Face Management improvements. I wonder whether the results of
this project are already integrated in digikam. Perhaps I overlooked it
in DK 5.x or DK 6 release notes,  so please forgive me to ask the same
thing again.

-- NeiNei

[Gilles Caulier-4]

The Face Management improvements using AI engine (deep learning) is integrate in digiKam core, but the code need finalization and improvements again. It cannot be yet used in production.

Gilles Caulier

2018-08-23 10:26 GMT+02:00 NeiNei <[hidden email]>:

Dear Gilles and team,

I just read about the DK 6 beta and remembered then GSOC 2017 Project 4 digiKam Face Management improvements. I wonder whether the results of this project are already integrated in digikam. Perhaps I overlooked it in DK 5.x or DK 6 release notes,  so please forgive me to ask the same thing again.

-- NeiNei

[PackElend]

Can I ask what kind of AI engine it is.

I cannot find any reference to OpenFace in the digikam documentation, why not using this one?

regards

Stefan

On 23.08.2018 10:49, Gilles Caulier wrote:

The Face Management improvements using AI engine (deep learning) is integrate in digiKam core, but the code need finalization and improvements again. It cannot be yet used in production.

Gilles Caulier

2018-08-23 10:26 GMT+02:00 NeiNei <[hidden email]>:

Dear Gilles and team,

I just read about the DK 6 beta and remembered then GSOC 2017 Project 4 digiKam Face Management improvements. I wonder whether the results of this project are already integrated in digikam. Perhaps I overlooked it in DK 5.x or DK 6 release notes,  so please forgive me to ask the same thing again.

-- NeiNei

[Gilles Caulier-4]

open face is Python library.

DK is C++/ Qt. it's not homogeneous.

We use Dlib neural network directly compatible with C/C++.

http://dlib.net/ml.html

Gilles Caulier

2018-08-23 15:36 GMT+02:00 Stefan Müller <[hidden email]>:

Can I ask what kind of AI engine it is.

I cannot find any reference to OpenFace in the digikam documentation, why not using this one?

regards

Stefan

On 23.08.2018 10:49, Gilles Caulier wrote:

The Face Management improvements using AI engine (deep learning) is integrate in digiKam core, but the code need finalization and improvements again. It cannot be yet used in production.

Gilles Caulier

2018-08-23 10:26 GMT+02:00 NeiNei <[hidden email]>:

Dear Gilles and team,

I just read about the DK 6 beta and remembered then GSOC 2017 Project 4 digiKam Face Management improvements. I wonder whether the results of this project are already integrated in digikam. Perhaps I overlooked it in DK 5.x or DK 6 release notes,  so please forgive me to ask the same thing again.

-- NeiNei

[PackElend]

Hoi Gilles,

thx for the update

Just checked OpenFace in detail and it says it uses dlib (or openCV) as well, so you have something in common :).

In case some is curious about the difference of the both, is well explained here https://www.quora.com/What-is-the-difference-between-dlib-and-opencv

On 23.08.2018 15:41, Gilles Caulier wrote:

open face is Python library.

DK is C++/ Qt. it's not homogeneous.

We use Dlib neural network directly compatible with C/C++.

http://dlib.net/ml.html

Gilles Caulier

2018-08-23 15:36 GMT+02:00 Stefan Müller <[hidden email]>:

Can I ask what kind of AI engine it is.

I cannot find any reference to OpenFace in the digikam documentation, why not using this one?

regards

Stefan

On 23.08.2018 10:49, Gilles Caulier wrote:

The Face Management improvements using AI engine (deep learning) is integrate in digiKam core, but the code need finalization and improvements again. It cannot be yet used in production.

Gilles Caulier

2018-08-23 10:26 GMT+02:00 NeiNei <[hidden email]>:

Dear Gilles and team,

I just read about the DK 6 beta and remembered then GSOC 2017 Project 4 digiKam Face Management improvements. I wonder whether the results of this project are already integrated in digikam. Perhaps I overlooked it in DK 5.x or DK 6 release notes,  so please forgive me to ask the same thing again.

-- NeiNei

[NeiNei]

Dear Gilles and Stefan,
thanks for the further clarification face management improvements (AI,
deep learning).

@ Gilles: Could you estimate whether the still necessary code
finalization and further improvement will make it in DigiKam 6 or a
later 6.x version, so that we could use the face tagging improvements in
production?

Thanks in advance,
NeiNei

On 23.08.2018 20:33, Stefan Müller wrote:

> Hoi Gilles,
>
> thx for the update
>
> Just checked OpenFace in detail and it says it uses dlib (or openCV) as
> well, so you have something in common :).
>
> In case some is curious about the difference of the both, is well
> explained here
> https://www.quora.com/What-is-the-difference-between-dlib-and-opencv
>
> On 23.08.2018 15:41, Gilles Caulier wrote:
>> open face is Python library.
>> DK is C++/ Qt. it's not homogeneous.
>>
>> We use Dlib neural network directly compatible with C/C++.
>>
>> http://dlib.net/ml.html
>>
>> Gilles Caulier
>>
>> 2018-08-23 15:36 GMT+02:00 Stefan Müller <[hidden email]
>> <mailto:[hidden email]>>:
>>
>>     Can I ask what kind of AI engine it is.
>>
>>     I cannot find any reference to OpenFace in the digikam
>>     documentation, why not using this one?
>>
>>     regards
>>
>>     Stefan
>>
>>     On 23.08.2018 10:49, Gilles Caulier wrote:
>>>     The Face Management improvements using AI engine (deep learning)
>>>     is integrate in digiKam core, but the code need finalization and
>>>     improvements again. It cannot be yet used in production.
>>>
>>>     Gilles Caulier
>>>
>>>     2018-08-23 10:26 GMT+02:00 NeiNei <[hidden email]
>>>     <mailto:[hidden email]>>:
>>>
>>>         Dear Gilles and team,
>>>
>>>         I just read about the DK 6 beta and remembered then GSOC 2017
>>>         Project 4 digiKam Face Management improvements. I wonder
>>>         whether the results of this project are already integrated in
>>>         digikam. Perhaps I overlooked it in DK 5.x or DK 6 release
>>>         notes,  so please forgive me to ask the same thing again.
>>>
>>>         -- NeiNei
>>>
>>>
>>

[Gilles Caulier-4]

2018-08-24 8:38 GMT+02:00 NeiNei <[hidden email]>:

Dear Gilles and Stefan,
thanks for the further clarification face management improvements (AI, deep learning).

@ Gilles: Could you estimate whether the still necessary code finalization and further improvement will make it in DigiKam 6 or a later 6.x version, so that we could use the face tagging improvements in production?

Wel at least Few weeks of full time to finalize the code, hack , and test. The last stage is the most important, as we need to be sure that no side effects are injected in the rest of the program.

Typically, the recognition data are in a separated database. So the errors will be limited. I think more about time computation, latency, crashes, etc.

I'm not sure also, if the data stored in database are well managed. The code is complex and need investigations.

Another question is about the code inside digikam core. i'm sure that we have too much duplicate and DLib shared library can be used as well in place. We don't have dependency to DLib. We take the necessery classes for testing quickly. As the code as good results in recognition, this point must be changed). In other words, the recognition with AI work but severals optimisations in source code need to be done.

The good point is that we have now 4 algorithms to process recognition. We have done comparisons, and Deep learning was the best.

But the problem is : it's slow. AI code in DK core is a demonstrator, we cannot use it in production.

This is why we don't turn on this feature for the moment. We need help to investigate. The human resources in DK is limited and this task require at least one developer to work on it.

We have more important priorities for the moment. So this feature is delayed, excepted if someone in this room is interested to work on it of course. You are welcome.

 

Best

Giles Caulier