Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Face.com API Alternative Beta Sign-Up (lambdal.com)
42 points by sabalaba on July 7, 2012 | hide | past | favorite | 41 comments


I know a lot of people who will be shocked by the announcement that Face.com will be closing their API. If anything, it was amazingly easy to hack a project together with face.com, not to mention it's academic state-of-the-art recognition.

We've been working on this alternative internally for some time. While it's still a private beta, we hope one day to provide the same level of service that Face.com did.


Open Source it otherwise I would believe you would just sell yourself to Google or Apple too.


To be perfectly honest, I sometimes wonder if the end goal of these acquisitions is to simply to acquihire all remaining talent in this space: PittPat (Google), polar rose (Apple), face.com (Facebook), etc.


Gah, not every product on the planet needs to be open source.


Great timing, I just came here to see if anyone would be able to suggest alternatives. My app Trollaroid is currently using the individual feature coordinates from the Face.com API (most importantly the sides of the mouth), does your service include this?


(I'm not Lambda Labs) I'm interested in perhaps making my face recognition and 3D reconstruction API more freely available. I've got a system that goes quite a bit past simple recognition - I'm able to identify facial features to a much higher precision than the publicly available APIs, as well as reconstruct a person from a photo, in photo-realistic 3D. Interested parties can visit www.cg-general-store.com/Auto-3D-Heads and/or email through my profile here.


The key benefit Face.com offered and what I'm interested in is linking the face to the facebook demographics. Anyone know of any face.com alternatives for this?


Pro tip: the 'email' field is not publicly visible on HN, if you want to publish your email here add it to your 'about' field. Stupid UI, I know.


Yes please. This would be great.


I was using face recognition. I'd really love to have that. Here's an idea: make the algorithm foss but have a paid super easy to use API. I'd pay money for that and I'm sure I'm not the only one! Just a thought on how you can get the best of being free and open while also monetizing.


Please do check our webpage at www.biometrycloud.com and have a look at our videos. We tested our own API over the past year or so, and now we are opening a private beta signup list. So far we have a few paying customers and want to expand quick


We've been thinking about doing this for some time -- potentially have a FOSS (host yourself) version and a paid cloud version.


1. Mood Detection was an excellent way to automatically update status.

2. Being able to Auto-Tag a Photo based on past knowledge of a person (whether celebrity or after training on general users) was a handy secondary feature, but admittedly this might be considered the most useful for some.

3. Last but not least, I always thought that being able to suggest similar photos (for example of products not just people) would have been a useful application in comparison shopping, but couldn't ever get that working well with Face.com's API.


Is there really a project here? Or are you just gauging interest in the wake of the face.com acquisition? According to whois this domain was created only a few days ago.


It's a newer domain because we used to be called 32ar (32ar.com registered 2012-03-23), I've been working on this full time since March.


We've been looking at the rage on twitter about this, please post what part of the face.com API you used below and we'll prioritize development:


Face recognition - attaching a facebook profile to a face. I was planning on building an entire business on top of that API feature. Luckily I'd procrastinated in this instance.


I was using the Mood, Smile, Male/Female, Age Estimation, Recognition and those are the key ones.


Facial recognition integrated with facebook was great, but the facebook part isn't necessary.


> the facebook part isn't necessary.

Necessary? Now it's the entire point.


I feel simple compared to all the others, but I was just using it for eye detection.


Facial recognition was the best out there and is something we're working on now.


ETA? Can I get a link or a beta signup?


Glad to see this. In addition to this, I wish there was a project I could contribute to that was focused on keeping Ruby/Pythhon bindings up to date for OpenCV...I mean financially, not in code, as I don't think I'd have much to offer. But even if I wanted to it's not clear how active the current maintainers are in responding to and evaluating pull requests


Not to mention that the bindings are essentially C with Python syntax as they are mainly implement their functionality by causing side-effects :/.


Hey just got the email after signing up.

I was using face.com's facebook facial recognition API, to match pictures taken with a webcam to somebody's facebook friends. I'm looking mostly for facial recognition/comparison, facebook integration would be nice but is not necessary.

I look forward to seeing what you guys develop. Good luck!


I wonder if Facebook will release an Open Graph API version of this? Or if they'll be data hogs and hold onto it? The former would help them leech more photos and perfect the recognition algorithms, but I guess with the amount of photos already in their system they've got enough data to play with.


Features ranked in descending order of importance (first entry being most important to me)

1) face recognition (with confidence value) [faces_recognize]

2) save tags/people to own namespace [tags_save]

3) get list of photos a person, or group of people are in [tags_get]

4) import feature to take face.com's discovered values

And for the service not to sold and then shut down.


I only wasted a few hours integrating face.com and used it for gender detection mainly. In the end a lot of african people showed as the incorrect gender so I'm okay with removing this option.

I am glad i did not use it further.

Face.com - you suck for shutting down


My company and I've been using the face.com face detection API extensively. OpenCV's in-built face detector just doesn't cut it, and we needed something built better.


Hi !, you could check our face recognition API at www.biometrycloud.com and if you are interested we are opening a private beta signup now !


Are you providing recognition or detection?


We are mainly interested in age and gender recognition in photos. Looking forward to your alternative API...


I used the mood, age, and sex api for a kinect project I was working on. Please sign me up.


I would like to see a good alternative for recognize method and group method.


Do you have any intention of open-sourcing this project?


Yes, eventually. The service might not be open-sourced, developing the algorithm as FOSS is something I've had in mind.


Wouldn't it be nice if we could get SLA's for APIs.


i was using face recognition and face detect with also age estimation in face.com

I hope you implement first that.


your about and team pages are broken


They're not broken, they're not linked/non-existent. They're using the '#'. Either the details will be anchored/bookmarked within the page or just a holder.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: