Hacker News Reader

Comment by gostsamo | Hacker News Reader

This tech could be a game changer for accessibility. Many card games are not or cannot be adapted to braille, so I haven't been able to play them with friends. If cards can be sold with invisible ink in an open format, then I would need only a scanner and a earphone to be able to check my cards without other help.

I've thought of doing with with QR codes, but I'm not sure if it will look okay for the other players and if the qr codes won't be recognizable enough by a human to give someone unfair advantage in some situations.

View on HN

Duplicate bridge cards are often printed with barcodes on the card faces, to facilitate automated dealing machines with computerised hand records. I've never seen a system which reads the barcodes for a blind player, but I'm sure it's possible. (The traditional approach is with a human 'card turner' assisting, which sucks for all kinds of reasons). Or even just reading the standard face images: it's a much easier problem than reading codes on the edges, and has the advantage of not making it easier to cheat!

Normal cards are a solved issue. You can print two braille symbols on each and it is done. Imagine though something like cards against humanity for example, where the text on the card would take ten times the space in braille. In different games images and details could have different meaning and simple ocr would be the wrong solution.

That makes sense!

If the QR codes or whatever are on the faces and only encode information already printed there (suit+rank for standard cards, the title and text for board games) I can't imagine there would be any cheating/unauthorised information problems.

Yep, this is the main idea at the moment. Currently looking for a game that needs the solution and is interesting for my social circle.

Dominion might be a good one.

Combined with a system like [0] on a head-mounted device with earbuds, it could make many card games accessible to blind and partially sighted people! No idea if this prototype was ever put into production though.

[0] https://www.youtube.com/watch?v=C0biAdmuons

Another option is an image recognition system trained on already existing cards. No need for special ink or qr code, just a camera as the reader.

Just OCR against the card-corners may be sufficient.

I guess you also need suits, but that's only four special characters (all of which may be in more-modern OCR-libraries).

As I said in another comment, normal cards are easy. The issue is with everything else where each card might have a picture with different animals and you are looking for the lama giving you the bird.

I think it is very doable for card games with thousands of different cards like MTG. I remember there was an old school SIFT/SURF image retrieval system that was able to deal with tens of thousands of album covers super fast. That was more than 10 years ago IIRC.

If the manufacturer is willing to provide the entire db of cards or a well-trained model that could be embedded or a ready mobile app, that would be great. However, waiting on them is not a winning strategy. The other option is community contributions, but you need a large community. This will be welcomed as well, but again not something that should be relied on. Both developments would be wonderful to happen, however in the meantime I'm looking for something cheap, fast, and easy to implement that I can do on my own in the local scope of my deck.

For some of the card games I am familiar with (Netrunner, MTG) there are resources online (wikis, dbs, deck builders) that have all the graphic assets. Alternatively, you can scan or photograph each card in your deck, but that may be time consuming.

As for software, there are engines out there that do content-based image retrieval (CBIR) out of the box [1]. It should be possible to build something quickly in OpenCV. You may be able to get away using simple image template matching by putting a few constraints on how the camera sees each card. Something more robust can be also be build using image descriptors and approximate nearest neighbors, as in [2].

[1] https://en.wikipedia.org/wiki/List_of_CBIR_engines

[2] https://blog.francium.tech/feature-detection-and-matching-wi...

There is already tech like this for TV poker games. RFID chips in the cards I think, and sensors in the table. There was a huge scandal a couple of years back about a player who had access to the info while playing. Google Mike Postle...

This is how a poker TV table actually work now. Each card has an HF RFID tag, each position has one antenna where players must put card when not seeing them. In addition there is another antenna in the middle of the table to read the common cards. In the past mini cameras were used for each position in combination with OCR to decode the cards. This sort of tables require collaboration from the players and from the dealer to obtain something useful to show on TV. They are instructed before the game.

Reply on news.ycombinator.com