Currently, one of the world’s biggest mobile computing conferences is underway in Barcelona, known as MWC. While the conference features the latest and greatest in mobile technology such as foldable phones and mixed reality headsets, there are rumblings of a bigger trend in the mobile sector.
These rumblings were first caused by Apple in 2017 when they first announced the iPhone X. Among other things, Apple had created a problem and solved it in a novel way with the launch of the phone. After pioneering fingerprint scanners on their devices for easy unlocks, Apple removed it with the X and replaced it with a complex algorithm known as FaceID. The company even thought it was secure enough to allow users to pay using FaceID as authentication.
This was achieved by a facial recognition model running on the phone and analyzing the data from 30,000 infrared sensors to match the 3D map of an individual’s face to other examples stored in the model. This offers it a fail rate of 1 in 1 million tries, making it suitable for authentication of financially sensitive information.
This was achieved using AI, necessitating the use of a dedicated inference chip on the board of the phone. Now, two years down the line, AI is quickly becoming prevalent as a game-changing feature for mobile computing, and is coinciding with the rise of other emerging technologies such as IoT, 5G and Big Data.
There is no better place to watch this effect occur than at MWC, where industry players are collectively innovating to change the state of the mobile market. Join us as we take a look into how the world of mobile computing is moving at a breakneck speed.
How Mobile Inference Creates A Foundation
This is one of the biggest talking points at MWC this year, with many players beginning to adopt on-board AI chips as necessary for higher-end smartphones. This is due to the fact that those phones usually come with a high-spec camera, leading the user to believe that the quality of the photos will be good. As the eventual average quality of photos go up, manufacturers are looking for ways to further improve the experience of taking photos on a phone.
This is the reason that many phones are going the way of the iPhone X and opting to include a smaller, secondary AI chip that analyzes the data captured by the camera at the time of taking a picture. While this is not used for something similar to FaceID, it is used instead to enable the facial unlock features on modern smartphones.
These chips are also used to conduct real-time tweaks on the face of the user in case of taking a selfie. In recent smartphones, the presence of an AI camera feature means that the user can change their skin tone, shape of facial features or even the size of their eyes by just tweaking options. The model will make the changes on the fly, allowing the user to tweak their appearance as they wish. However, this is just one of the uses of AI chips on smartphones.
How The Industry Will Move Forward
The advent of technologies such as 5G has created new use-cases for on-board inference in AI-enabled devices. With the term ‘edge computing’ becoming more and more widely used as time passes, the definition of the term has begun to vary in order to reflect the state of real-world developments. The one thing that is certain for the time being is that cloud computing devices have latency, thus requiring on-device solutions for providing day to day functionality such as unlocking phones.
However, a hybrid approach with both an on-device chip and the power of an always-connected cloud with the use of 5G will lay the foundation for the creation of novel user experiences. With the current market continuing to innovate on top of existing improvements, the paradigm shifts that occurred in the early heydays of mobile computing are yet to be replicated. However, as the market continues to build the infrastructure for the next step of advancements, change is sure to come.
With the industry’s biggest players gathering in one place, we are bound to see the signs of the coming future. Indeed, there is a huge push for 5G, unsurprisingly, by many prominent telecom service providers and smartphone manufacturers. 5G is said to be how the world will be connected in the future, and is set to offer a substantial boost to cloud connected services such as what we see today. Moreover, it can also enable the rise of the Internet of Things by facilitating such features as machine-to-machine payments and an easy interface between mobile devices and the cloud.
Cloud programs will be able to perform almost real-time actions when 5G is deployed, allowing manufacturers to cut down on the number of components required to be inside the phone itself. Innovations such as foldable screens and phones with dual screens will also set the stage for the face of computing to change with the advent of faster networks.