Developing a facial recognition enabled product to dispense toilet paper? A quick analysis of the Amazon, Google, Microsoft, and IBM facial recognition platforms.
As a park in Beijing starts using facial recognition to dispense toilet paper, our imagination about possibilities started running wild.
Facial recognition systems have commonly been used for security purposes but are increasingly being used in a variety of other applications. Some mobile payment systems use facial recognition to securely authenticate users, and facial recognition systems are currently being studied or deployed for airport security. According to a report from CBS News, almost half of United States citizens are represented in a facial recognition database. Facial recognition is also supposed to experience increasing usage as Apple and Samsung are planning to integrate it, in their new models. Samsung phones have used facial recognition for unlocking devices in the past, but in their latest Galaxy S8, they extend the technology to support secure financial transactions. Apple is also paying attention to facial recognition by acquiring a patent for a way to detect faces using information in digital video feeds. (Bloomberg reports)
Picking the right platform requires research to understand the best fit. It’s not just product features that help you choose your facial recognition software partner instead things like price, reliability, performance, reputation, support, and ease-of-use can all be deciding factors. With some of the biggest players (Google, Amazon, Microsoft, and IBM) rolling out their own offerings, it’s an exciting time for the market as the cost of the end product should decline while options should continue to improve. Below is a quick comparison among these four service providers based on the core features provided by them.
Deep learning based image recognition which is part of Amazon’s AWS Ecosystem. The platform is based on their acquisition of
- Object & scene analysis.
- Face detection.
- Face recognition.
- Face sentiment analysis.
This platform is strong in content analysis, as it classifies images and video into thousands of categories. This is marketed as part of Google’s Cloud Platform.
- Detect inappropriate content.
- Image sentiment analysis (faces).
- OCR & automatic language identification.
- 'Video Intelligence' - Scene, object and 'entities'.
Microsoft Face API provides image and video analysis on faces. It is marketed as part of Microsoft's Cognitive Services Platform.
- Face detection.
- Face verification.
- Face identification (recognition).
- Emotion detection.
The platform understands the content of images and is marketed as part of the Watson Developer Cloud.
- Face detection.
- Celebrity recognition.
- Object & scene recognition.
Each player has approached the facial recognition challenge using different algorithms and core features. Your choice should be dependent on what you are trying to achieve for your product or business. If you are interested to know more about Facial Recognition APIs and wondering which one to chose and how to use them for your product, feel free to contact us at email@example.com - - we'd love to hear from you.
Kaggle has a bit of a history with Google. Earlier in March 2017, Google and Kaggle teamed up to host a $100,000 machine learning competition around classifying YouTube videos...
Asset Optimization is a structured approach to improving the effectiveness of overall asset management by classifying, defining, scheduling and implementing initiatives for better performance and decision making...
Google is also developing another deep learning algorithm that surveys biopsy images to locate metastatic breast cancer that has spread to the lymph nodes. The algorithm was tested...