Re: [Computer-go] [ANN] Imago - Go board optical recognition

[Mario Corsolini]

Hi everybody!
Two days ago we released PhotoKifu v3.00, that heavily relies on OpenCV
instead of the nice, but slow, ImageMagick suite. We are now able to process
each picture in about a tenth of a second, pre-processing included.
Everything may be found at http://www.oipaz.net/PhotoKifu.html, including a
new demo video, this time featuring the Guo Juan vs. Crazy Manja game,
played this summer in Liberec during the Second International Go Game
Science Conference. This was an important game from many points of view, as
not only a professional player (Guo Juan) was involved, but a test was
performed on the advantages of human-computer cooperation, with full success
(the human-computer team performed beyond expectations for human alone).
This demo video is available at
http://www.oipaz.net/REP/GuoJuan-CrazyManja.mp4. All the pictures from the
game are at http://www.oipaz.net/REP/GuoJuan-CrazyManja.zip.

We'll probably release one more version of Photokifu, then we'll start
working on VideoKifu, a (very) preliminary version of which we hopefully
will be able to demonstrate next summer, during EGC2016, although at the
moment it's not certain that a scientific conference will be held in
Saint-Petersburg too.

-- 
Dr Mario Corsolini

___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

Re: [Computer-go] [ANN] Imago - Go board optical recognition

[Mario Corsolini]

Hi Josef, and everybody!

I am Mario Corsolini, the other author of PhotoKifu. I am aware our program
was a little hidden in the Internet, mainly because until few weeks ago only
an Italian version existed. Now you can download the international (EN/IT)
version from my website:

http://www.oipaz.net/PhotoKifu.html

Within few days v2.10 will be available, featuring an improved recognition
routine (as previously stated by Andrea), as well as various bugfixes.



After the tests in Pisa we will be able to provide more details, re its
reliability. For the moment I just want to point out that PhotoKifu still
requires a human operator: pictures must be shot during the game, then
transferred into a computer and analysed by our program.

In order to achieve a higher level of automatism we must complete VideoKifu,
which at present is in a very early stage of development. I suppose in
ECG2015 we will have a working prototype for a demonstration, at best: an
official release will most likely be ready before ECG2016.

Best regards,
Dr Mario Corsolini

___
Computer-go mailing list
Computer-go@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go