In a previous blog, I introduced the SLAM technology, explaining what it is and how it works. As I mentioned, I believe that SLAM has the potential to be truly transformative across a number of different devices and technologies from mobile to autonomous vehicles. I also believe that there will be five genuine areas where SLAM will improve and fundamentally change the lives of consumers and businesses in the near future. These are navigation, advertising, retail, gaming and social, including social media and day-to-day interactions with friends, family and even fans online.
Let me take you through these use cases in more detail and explain how SLAM could transform each one.
For navigation, SLAM will improve the accuracy of outdoor navigation, while also enabling accurate indoor navigation. Just think about being able to navigate around a giant shopping center or any large indoor arena on your smartphone without having to ask for directions or look at a physical map. So how does this work?
SLAM will bring a far greater understanding of the physical environment - both indoors and outdoors - so that enhancing it with virtual elements becomes possible. As SLAM brings position and mapping information to the software level then developers can utilize this information and overlay the environment with graphics to help users navigate to their destination. This could be achieved through some arrows drawn on top of a street or an indoor corridor, or even through virtual characters guiding users to their destination. Personally, I cannot wait to see how it will impact future tourism. I can already imagine being guided through a city with my children's favorite cartoon character providing entertainment for the youngsters.
SLAM is already in millions of phones thanks to Google ARCore and Apple ARKit. A previous blog from my colleague Roberto Lopez Mendez explains how to implement indoor navigation using Google ARCore. In addition, this short video below shows just how it works.
SLAM navigation holds the key to advertising, with the two being developed hand-in-hand. In fact, advertising is possibly one of the most exciting SLAM use cases from a business perspective. SLAM will be able to provide an extremely accurate location of the user, as well as the direction that the user's smartphone camera is facing. Therefore, the digital map provider will be able to establish exactly where the user is located with no errors or miscalculations and exactly what the user is looking at, such as a particular shop or building. Once this is established, the advertising will kick in, providing a targeted advert to that user. For example, if SLAM works out that you are looking at a sports shop then the device will ask whether you would like to buy certain products from that sports shop or offer a ten per cent discount for the next 30 minutes. The accurate targeting of adverts to users based on their location and what they are looking at has the potential to be a game-changer for retailers. In theory, a mobile device could work out what the user wants to buy even before they have thought of it!
For retail, SLAM will allow consumers to use their smartphone to transpose a physical object virtually into the physical environment, such as the person's living room in their house. Just imagine using a smartphone to take a photo of a wardrobe and then transposing it virtually into the bedroom. Consumers will be able to see whether the wardrobe will fit without needing to take any measurements at all. The ability of SLAM to provide very accurate measurements means that this is possible, with the technology working out the precise dimensions of the environment and the object.
SLAM could not only transform physical shopping, but also online shopping too. Online retailers could use the technology to provide a 3D virtual object against real world metrics. Consumers shopping online can then drag and drop the virtual object to see if the item fits in the room. If the item fits in the virtual world, then users can rest assured that it will definitely fit in their space. Therefore, SLAM is not only going to help consumers buy the right items, but also reduce the amount of products that are returned.
For gaming, SLAM will provide the environmental context to make games 'on the go' - often on mobile - even more meaningful and exciting. Using Dense SLAM is likely to be the ultimate goal for gaming, as the semantics will give developers huge freedom to express their creativity to engage end-users. That being said, today's Sparse SLAM is already a massive leap forward compared to what developers had before. If they could provide an engaging and universally popular game such as Pokémon Go without SLAM, just think of the possibilities with the technology. This potential in gaming could be a truly exciting development for mobile devices.
SLAM can also be used for tracking the user's position in VR and mapping it to the virtual world. A previous blog explains how to implement mobile inside-out VR tracking on our phones using Google AR, while the video below shows how it works. In the near future, we should be able to play new types of games on our mobile phones that are currently only possible on console and PC. The new generation of all-in-one VR headsets based on mobile SoC are already using this technique, with some already released (e.g. the Lenovo Mirage Solo).
Finally, on social, SLAM opens up new possibilities for developers to change the way people use social media and how they share their experiences. Social media is quite a broad term and it is hard to quantify the number of use cases where SLAM would be applicable. However, there are few obvious potential examples, such as wearing a status above your head as you walk and live performances with virtual characters. Another example that would be made possible by SLAM technology is turning a room into a virtual TV studio for live broadcast performances, which presents a range of opportunities on video sharing platforms, such as YouTube. This has the potential to turn fairly amateur YouTube vlogger productions in someone's living room or bedroom into something that looks far more professional.
Despite these truly exciting use cases, there are a number of system challenges associated with SLAM. It will need to synchronize data from camera to SoCs to display. SLAM will also require new calibration, with camera vendors needing to provide Android interfaces. For indoor navigation, the main challenge is synchronising a starting user position with the position in the map; however, SLAM represents the best technology to overcome this particular challenge. Finally, there is currently a lack of common standards for SLAM, which could be challenging with so many different requirements for the technology. These challenges will need to be addressed if SLAM is to progress as a viable future technology across a number of use cases, not just the ones highlighted in this blog.
Overall, we believe that SLAM will be a long-term growth opportunity for the technology and business sectors. We are encouraging our Ecosystem of partners to work together to provide cohesive solutions for the whole system. Our message is to join Arm and embark on this journey together. If any developers are working on exciting SLAM solutions, then I would encourage them to get in contact with me at sylwester.bala@arm.com
I'll be presenting more about SLAM in my TechTalk at Arm TechCon on Thursday 18th October.
[CTAToken URL = "https://www.armtechcon.com/registration.html?utm_source=arm&utm_medium=blog&utm_campaign=2018.7_techcon_techcon_na_na_na_me2767a1_00" target="_blank" text="Register for TechCon" class ="green"]