Online AI services; the two ends of the spectrum differ significantly. At one end there are infrastructure solutions which grant more control and customisation. Whereas at the other, we have pre-built API services. Here at Omobono, we’re finding that many solutions now involve both of these, and over the past year, we’ve been using more service APIs. Here’s a breakdown of some of our favourites.
Broadly they’re divided into these areas of implementation:
Image and facial recognition
Having been on the market for less than a year, True Face has already established itself as a leader amongst similar offerings with a range of excellent features like spoof detection and blurring of unknowns. Its beta also sports nice features around gender and age detection.
Karios has a useful feature set which includes, gender, age, attention, emotion, ethnicity and facial matching.
A great fit if you are an AWS user with features like suspicious content recognition, text in image detection, face comparison and emotion detection. Amazon Rekognition is very easy to use.
4. Lambda Labs
Face detection and facial feature location.
Besides facial recognition, it has a range of number plate, car make and model recognition models as well.
A straightforward and free API for face detection.
Mature facial recognition API
Main features include:
The face detection by Skybiometry is also a reliable solution that allows for the detection of people with or without glasses, captures multiple faces in the photo and analyses their emotion.
To analyse and recognise faces more efficiently, this API can transform a 2D picture into a 3D model. The recognition is performed by detecting a face and comparing it to an existing set of faces until the match is found.
10. Face ++
This API can be used for online advertising and is fully featured with detection, landmarking, comparison and searching. It also has an offline SDK for iOS and Android. It can provide facial detection and comparing even when a user’s phone has no reception. This gives a broader range of possibilities to developers.
With many of the same features as Google Cloud Vision and IBM Watson, including celebrity detection and OCR. An exciting piece of this API is that in addition to listing tags along with confidence predictions it also attempts to generate a natural-language description based on those tags.
Upstart image recognition service that also uses a REST API. One interesting aspect is that it comes with a number of modules that help tailor its algorithm to particular subjects, like weddings, travel, and food.
Taking a slightly different approach to image recognition. CloudSight’s website offers significantly fewer details than some of their competitors. Providing a combination of algorithmic and manual (meaning human) tagging.
Google’s visual recognition API based on the open-source TensorFlow framework and using a REST API detects individual objects and faces. It contains a pretty comprehensive set of labels and also comes with a few bells and whistles, including Optical Character Recognition (OCR) and integration with Google Image Search to find related entities and similar images from the web.
Machine Learning and predictability
1. Big ML
BigML provides a selection of Machine Learning algorithms by applying a standardised framework.
BigML covers not only classification, regression and time series forecasting but also unsupervised learning tasks such as cluster analysis, anomaly detection, topic modelling and association discovery.
An E-commerce recommendation engine.
MLJAR provides a selection of Machine Learning algorithms that you can tune and deploy models as services to the cloud.
NLP, text analysis, and sentiment analysis
An NLP API that provides support for bot building, customer feedback analysis and general NLP tools. It also includes features like Lemmatization, Post Tagging, Language Detection, Phrase and Entity Extraction and Sentiment.
2. Datum Box
This API provides, Sentiment Analysis, Topic Classification, Language Detection, Subjectivity Analysis, Spam Detection, Reading Assessment, Keyword and Text Extraction
With features like Bing Spell Check, Language Understanding (LUIS) Intent recognition using NLU, Text Analytics and Translator Text. This is an API we’ve used extensively to build bots for clients.
Language translation and recognition
Automatic Speech Recognition (ASR) and Natural Language Understanding (NLU) all in one engine. This is used a lot in automotive and OEM industries.
Speech to text, Text to Speech, translation across 60 languages, Speaker recognition and a custom speech service that you can train.