Artificial intelligence (AI) is the property of intelligent systems to perform the creative functions that are traditionally considered the prerogative of man. Currently, the development of this area is one of the most promising. By itself, AI is used in various fields of science and technology, but the article will be considered a small area of its application, namely – smartphones.
At the end of 2017, the company ARM introduced a new generation of processors with the microarchitecture Dynamiq. The developers have positioned their ARM chips as those that will be easier to tune. This allowed manufacturers to create more powerful systems designed not only for conventional devices, but also for the decision of technological problems of the future — artificial intelligence and unmanned vehicles.
Dynamiq can adapt to the user — while in standby mode, it uses the core with less power consumption, and to download the application or game uses a more powerful processor core. Dynamiq also creates special conditions for work with artificial intelligence to provide improved performance 10 times.
These functions, as expected, used in modern smartphones with built-in AI. But there were more interesting field of use.
Camera smartphone identificeret different types of objects in real-time to determine the scene that will help inexperienced photographers to make good photos in the “got took”. The neural network optimizes the color, contrast, brightness, exposure and other advanced camera settings when recognizing pictures. At the moment, the AI in the smartphone can recognize up to 13 different objects (human, animal, plant, landscape, etc.). In the future this list will be supplemented.
Thanks to intelligent algorithms, the quality of photos taken with a smartphone, compared with almost professional photographs, processed in graphic editors.
The use of machine learning for system optimization
This technology ensures that your smartphone will be able to work equally fast regardless of whether it was used only hours or a few months. Huawei Mate 10 Pro is equipped with CPU Kirin 970 and intuitive interface that provides shell EMUI 8.0. Smartphone learns the user’s habits and adapts to them to optimize performance. Phone identificeret applications that are used more often and stores them in RAM, which speeds up their activation. A similar process occurs when the resource management processor: the most popular apps are using the fastest CPU core. Less critical applications are removed from memory to reclaim space.
The real-time translation
Thanks to AI translation of the text is 300% faster and takes place in offline mode without Internet access. Huawei is able to translate with almost 50 languages, and so-called “Conversations” allows in real time to communicate with people from other countries.
AI in the photo gallery
Algorithms for AI to help you create a intelligent library photos. Huawei automatically tags photos, identificeret images, settings, people, and location photography.
The developers of Huawei have been able to find other uses for AI in the smartphone. It was directed at the control of the vehicle and the object recognition on the road. Another example of the fact that the can be used in different areas.
Learn more about the smartphone: a Review of Huawei Mate Pro 10 – a smart flagman with the AI
The company Samsung has in its Arsenal of intelligent voice assistant Bixby, which is used in smartphones Galaxy S8 and Galaxy Note 8and the new products – Galaxy S9 and S9+. Unlike Apple Siri, which primarily is a search engine for voice commands, Bixby can identify items through the app for smartphones.
With Bixby Vision there is an opportunity to learn about the objects and places using the camera. Voice assistant lets you know how to buy food, can translate into different languages, and perform many other tasks. Bixby Voice is available for smartphones and is able not only to retrieve information but also to perform voice commands such as opening an app or set an alarm.
AI in the Apple focuses on image and animation, and implemented in iPhone X, equipped with the new six-core chipset A11 Bionic, in which there is “neural system”. NPU consists of two processor cores that control “special machine learning algorithms”. Thanks to them, the implementation of the technology Face ID, Animoji and support for augmented reality. Built into the smartphone of the neural network is able to perform 60 billion operations per second.
The introduction of the “neural system” allows the hardware for AI is to the smartphone, the company sends less data to the server and can better protect the privacy of users.
More detailed about the smartphone: the iFixit Team has already dismantled the new iPhone X
Google also decided to use AI for photos and video. Here we will focus on the image processor. It’s called Pixel and implemented Visual Core OCTA core CPU which is used in Google and Google Pixel 2 Pixel 2 XL. SoC Qualcomm Snapdragon 835 contains its own image processor, but Google apparently felt that was not enough.
About Pixel Visual Core know quite little. The CPU performance is more than three trillion operations per second, allowing you to process images in HDR+ mode five times faster than before. Pixel Visual Core provides the ability to capture the signature HDR+ mode in any camera app. The platform itself is a system-on-chip, not just CPU, since its configuration includes the Cortex-A53, EN RAM LPDDR4, and several other blocks.
Learn more about smartphones: Google Pixel 2 and Pixel 2 XL officially presented
The company decided to use AI software Amazon Alexa. The main platform for the Alexa – smart speakers Echo and Dot from Amazon, turn off the light, reading books and eating the goods at home. They collect data about their users and improving the process, gaining new skills and becoming more aware of what they want from the user. There are also apps with “smart” voice assistant for smartphones running Android and iOS. Basically Alexa is used for the “smart” house.
At the moment, Alexa is able to:
- close the door, control the light, TV, air conditioning and other systems of smart home;
- timer or calendar;
- to tell you the latest news;
- to play “rock, scissors, paper” or flip a coin;
- to buy goods, order a pizza, buy tickets to the cinema or on a plane;
- read, translate unit;
- play user quizzes;
- to consider how much time is left before the New year;
- finding a smartphone in the house (apartment);
- To tell how much fuel remained in the car and turn the engine, if the user wants to go outside (only for BMW);
- to read audio books or lullabies;
- order/cancel Uber and Lyft;
- to calculate how many calories are in the food;
- to answer questions (depending on the context – or with humor, or in the style of “Wikipedia”).
Moto Force Moto Z2 and X4 , the first smartphones that use the platform neural processing Neural Processing Engine for the artificial intelligence. Using Qualcomm Snapdragon NPE, the developers of Motorola Mobility has added to smartphone cameras feature Landmark Detection that can identify more than 1,200 facilities around the world with the help of neural processing. The program works in beta mode, and the object list is constantly updated.
This feature is especially useful for travelers. And even if hometown seems familiar, object recognition using neural networks will help to take a new look at familiar structures.
V30S LG ThinQ smartphone which uses artificial intelligence. This is the first smartphone of the company that is part of the family of ThinQ, which is a device with built-in artificial intelligence (AI). Smartphone use Vision AI associated with the camera and AI Voice (recognition of voice commands).
Vision AI includes AI CAM (the system analyzes the object in the scene and recommends the best shooting modes for a given object), QLens (after scanning a QR code, the smartphone will display information on where to purchase it cheaper and will offer similar products; search from the camera to the original images to obtain information, e.g., about minutes) and Bright Mode (improves the quality of the photos taken in low light – new algorithms give the opportunity to make a photo twice as bright and no strong noise).
Voice interface Voice AI is working on a couple of Google Assistant. The owners of LG 32 V30s will be available exclusive voice commands that do not exist in the solution from Google.
The company implemented AI features for cameras and overall performance of smartphones ZenFone ZenFone 5 and 5Zpresented on MWC 2018. In both models, technology support, ASUS AI Boost, which is designed to increase performance in games and other applications that require high computing power.
With regard to artificial intelligence in the cameras, here we can distinguish the following 4 main functions:
- AI Scene Detection (detection of 16 different scenes and objects).
- AI Photo Learning (learning user preferences during shooting).
- AI Real-time Portrait (portrait in real time).
- AI Real-time Beautification (improving self-portrait in real time).
Learn more about smartphones: ASUS ZenFone 5, 5Z, 5 Lite and ZenFone Max (M1) presented at MWC 2018
Prospects of development
Analysts with the company Counterpoint Technology Market Research predict a gradual proliferation of smartphones with support of technologies of artificial intelligence (AI) and machine learning, embedded on the processor level.
Thanks to the technologies of artificial intelligence and machine learning smartphones already have the ability to perform a variety of complex tasks, including translation into foreign languages in real time, photos and video in high-quality thanks to the automatic adjustment of the optics under ambient conditions, etc.
In the future, it is expected that smartphones with AI will be able to act independently and to provide users only the information they need based on interests and habits. It is difficult to imagine what areas will be used in smartphones of the future. If you are already a smartphone with AI can drive a car, then possibly in the future, its application will affect production processes.
Analysts predict rapid proliferation of smartphones with artificial intelligence in the segment flagship devices, and by the middle or the end of 2018, the AI get to the middle class. Forecast Counterpoint stipulates that in 2020 more than a third of all supplied to the market of smartphones will have integrated AI capabilities.
The development of is an interesting and promising direction. Despite the fact that the introduction of new solutions based on AI requires a lot of investment and knowledge in all fields of science, the effect of this technology should make human life easier. The AI in the smartphone performs the role of an assistant who will attend to itself, the user has more free time for other things.