@@ -11,7 +11,7 @@ Therefore our current research focuses o
### Sansa ###
First detector approximately locates the board in the picture. It uses SSD (single shot detector) with MobileNet and is working decently.
Examples (downloaded from Flickr under CC0 license):
Examples (downloaded from Flickr under CC0 license):


@@ -19,10 +19,10 @@ Examples (downloaded from Flickr under C
### Dochi ###
Then the cropped and normed image is fed to the second detector, designed to return precise grid corners' coordinates. This is currently implemented by a convolutional neural network and obtaining the required precision is a major hurdle to be passed.
Pretty good:
Pretty good:

Also good, but not enough:
Also good, but not enough:

### Genjo ###
@@ -35,12 +35,12 @@ Base case: we have two correctly recogni
Issues:
* illegal positions → ignorable
* illegal positions -> ignorable
* positions unreachable from the previous state
* reachable from any past state. (Incorrect states inbetween.) How to pick the correct leaf of such a tree?
* reachable by more than one move. (Lost states inbetween.) Issues with branching factor.
* stone shifts
* stone stops being recognized → fixable manually and even ignorable
* stone stops being recognized -> fixable manually and even ignorable
* stone is recognized at an empty intersection. It can be occupied later for real. What do?
This is handled by the `statebag` package and its documentation also states the problem and used solution in precise terms.