AI/Machine Learning/NLP/AR/VR 2018-03-13T16:28:30+00:00
Quantilus AI NLP ML

Artificial Intelligence, Machine Learning, Natural Language Processing, Augmented and Virtual Reality

Intelligent Automation for Work and Life

At Quantilus, we have been working with AI before it became cool (and scary). Our first foray was in the field of Natural Language Processing – which we used for automated grammar and style checks of written content. Subsequently, we built tools to classify untagged content in intelligent, usable ways, and to present it for consumption with a high degree of personalization. More recently, we have been working on personality assessment of individuals based on 1) the words they speak (a relatively simple task), and 2) changes in their facial patterns based on verbal and visual cues (a much more complex task).

Want to build Virtual Reality or Augmented Reality apps for your business? We built some of the first business-focused AR apps for mobile and wearable platforms through our SAP partnership. Our apps help technical support personnel visualize product models, and also let customers visualize retail products in empty space. With the added complexity of tight integration with backend ERP systems.

FEATURED WORK

Appliqant - Automated Video Interviews
APPLIQANT – the Automated Interview Robot. Disrupt recruitment through the automated screening of job candidates.
Appliqant - Automated Video Interviews
Wearable Apps for Technical Support

Discover Simple Assist – Wearable Apps for Technical Support

Wearable Apps for Technical Support
Visual Showroom - Augmented Reality

Visual Showroom – Augmented Reality App for Product Display

Visual Showroom - Augmented Reality
Deloitte - Automated Document Editing, Data Extraction

Deloitte – Automated Document Editing, Validation and Data Extraction

Deloitte - Automated Document Editing, Data Extraction
Intel - Machine Learning

Intel – Machine Learning for Customer Classification and Segmentation

Intel - Machine Learning
Intelligent Lease - Data Extraction using NLP

Intelligent Lease – Automated Data Extraction from Unstructured Lease Documents

Intelligent Lease - Data Extraction using NLP
BluePencil - Grammar and Writing Style Checks using NLP
BluePencil – Automated Grammar and Writing Style Checking using NLP
BluePencil - Grammar and Writing Style Checks using NLP

TECHNOLOGY STACK

Some of the frameworks and tools that our development teams have used recently. A list that grows by the day.

RELATED RESEARCH

Relevant, interesting and current curated research content in the field.

The Science Behind Deepfake Technology

March 5th, 2021|Categories: AI, NLP, Machine Learning, Emerging Tech - AR, Cloud, Blockchain|Tags: , , , , |

Recently Tom Cruise probably woke up to find trending videos of himself playing golf and performing magic tricks for TikTok that he has no recollection of ever recording. But, unlike a plot out of a Hollywood blockbuster where the main character struggles to remember their murky past, Tom Cruise never actually recorded these videos.

Unfortunately for Tom Cruise, he’s the latest, viral “deepfake” and probably nowhere near the last. This quickly expanding technology has seen a growth in attention, possible future use cases (both useful and potentially harmful), and discussions surrounding ways to regulate the technology.

How Does Deepfake Technology Work?

At the heart of deepfake technology are machine learning and AI. For example, to create an uncanny replication of Tom Cruise, the videos’ creator would have trained a neural network to identify what Tom Cruise would look like from every angle or in different lightings and environments. Once the AI can generate a “face,” a user can superimpose the deepfake onto any media or another actor just as easily as someone working with photoshop.

Also, one reason why Cruise or any celebrity/public figure, in particular, are easier targets is that there are countless photos and videos that can cover every angle a machine learning network needs to piece together a recreation.

The rise of this technology mainly has to do with these projects becoming cheaper to run, and the AI behind this technology is becoming faster and faster. That’s not to say anyone can start creating their own personal Tom Cruise as this process still takes plenty of hours and concentration to adjust every little detail that may hinder the end result’s realism.

Another aspect to keep in mind is that the deep-learning technology and algorithms behind the video deepfakes of Tom Cruise differ from the fake avatars that are becoming more common across social media. The technology that The New York Times focused on is powered by generative adversarial networks (GANs). As the article mentions, there are still tiny flaws that are noticeable as new avatars are created, and GAN models specialize in creating imagery instead of video.

Where is Deepfake Technology Headed?

Forensic experts and those who analyze digital images can still determine what is and what is fake. However, that threshold is becoming harder to distinguish. For the Tom Cruise example, Hany Farid, a professor at the University of California at Berkeley (Cal-Berkeley), says the videos are well done.

Farid, who specializes in analyzing digital images, said his graduate students could identify Cruise’s eye color and shape change for a brief moment at the end of the magic trick video. So, for the moment, AI technology isn’t perfect.

It’s also suggested by Farid that, in the case of the Tom Cruise videos, part of the realism came from the actor’s mouth, though the rest of the face was generated by deep learning: “This would make sense if the actual person in the video resembles Cruise, did some good work with makeup perhaps, and the swapping of the distinct eyes is enough to finalize a compelling likeness.”

For the time being, there’s still a precise skill and monetary gap before anyone can start creating their own AI-powered deepfakes, but that timeframe is gradually shrinking. There’s not much of a consensus of when this technology will become refined and easily accessible. Still, some experts suggest the range is two to ten years before it can be as easy as pulling up an app on your smartphone.

Unfortunately, the shrinking timeframe also means less time for companies and even governments to put forward solutions and regulations that can help reduce some of the harm and weariness that can come out of AI-powered deepfake technology.

How Can Deepfake Technology be Regulated?

Large technology companies like Facebook, YouTube, and Twitter understand their platforms are commonplace in the modern-day news cycles and have started to reduce the threat that all deepfake technology can impose.

Facebook recently recruited researchers from Cal-Berkeley and Oxford University to build a detector that can identify and remove any deepfake content that can be harmful. Twitter also has plans to tag any deepfake that isn’t initially removed. And YouTube prioritized removing deepfake videos that could have impacted information surrounding the 2020 election in the United States.

There are also deepfake analyzing programs such as Reality Defender that can act almost like a spam remover works on any emailing platform and alert users if a piece of media may be doctored. The company is collaborating with Cal-Berkeley, Twitter, Microsoft, and Google.

Lastly, another emerging technology and fellow buzzword of the past few years, blockchain, can potentially provide a solution to hinder unregulated deepfake content. Blockchain is now capable of utilization for creating watermarks on pieces of digital content.

For example, an original video created by a celebrity or public figure can have a generated unique signature that proves that the piece of content is original and un-doctored. That watermark can act as a way to debunk deepfake content or misinformation quickly.

What’s Next for Deepfake Technology?

The Tom Cruise deepfakes are ultimately accelerating a necessary discussion about a technology that is becoming more and more prominent. There are also some positives of the growing technology for all the potential threats of deep learning AI. Image-based media isn’t its only purpose for the world. From innovating healthcare to art to education, deep learning AI shouldn’t just ignite fear but rather encourage digital literacy among the public so that they can identify a deepfake from reality.

Contact us at info@quantilus.com for a consultation and learn more about what Quantilus has to offer here.

The Best of Emerging Tech from CES 2021

January 22nd, 2021|Categories: AI, NLP, Machine Learning, Emerging Tech - AR, Cloud, Blockchain, Mobile Development|Tags: , , , |

Last week we showcased some of the best product innovations generating buzz out of the virtual Consumer Electronics Show (CES) this year. While products usually make the highlights of the show, the spotlight shines brightest on the technology that powers them.

Most illuminating is how emerging technology evolves in the context of unchartered, innovative applications. Emerging technology trends such as artificial intelligence (AI), machine learning (ML), the internet of things (IoT), and blockchain is behind some of the game-changing tech that could impact life and work in the near future.

Voice Technologies

Voice technologies are one set of AI and ML that is constantly providing more ways for people to communicate with each other. For example, mobile translation apps or Google Translate are able to help users navigate a different language in almost any situation. However, usually, these apps can’t always handle translating a full conversation correctly.

TimeKettle announced M2 Translator Earbuds at CES 2021. The software utilized for these earbuds differs from the normal translation app because each person with one earbud can select their desired language and engage in a full conversation.

The software also has Voice Activity Detection (VAD), so voice pick-up and translation are completely automatic. TimeKettle also utilizes Bluetooth 5.0, updates the earbuds for more languages or new features on its own, and translations can occur in areas without an internet connection.

In terms of how much space the voice technology can hold, the earbuds provide access to 40 languages, over 90 accents, and 12 servers. By TimeKettle’s estimations, they give translation access to 95% of the world population.

In addition, the earbuds have adaptive noise cancellation and use an algorithm to improve the quality of anyone’s voice input. Lastly, the product can also act as headphones for phone calls and music. The M2 Translator Earbuds are currently available.

The voice technology TimeKettle announced is another example of a technology that can help remove language barriers for businesses. And consumers shopping in different countries could even become more active shoppers when being able to start a conversation with a business owner who speaks a different language.

AI-Powered Diagnostics

MediWhale announced a diagnostic device powered by AI and it can predict eye diseases and cardiovascular diseases in just a few minutes.

The health monitoring device is powered by AI that uses an ophthalmic camera in addition to medical records. This device analyzes images and makes decisions and diagnoses by comparing images with an integrated database of systemic diseases.

One benefit of the “DrNoon” device is that it’s a cheaper and faster test to run than the average than a heart CT. Also, there is no common risk of radiation exposure that a CT might have.

MediWhale wants to provide a new biomarker for heart disease: “Our goal is to analyze the fundus image through AI and to present the fastest way to detect the disease through it.”

MediWhale claims “DrNoon” can achieve a 95 percent accuracy in identifying eye diseases and 84 percent accuracy for cardiovascular diseases.

The quick diagnostic technology MediWhale is working on can ultimately lead to hospitals or urgent care centers have more resources to use when dealing with an influx of patients.

Machine Learning Cybersecurity

ArcRAN announced at CES 2021 a machine learning algorithm that can help enterprises and governments to quickly respond to potential cyber threats and attacks. ArcRAN has pushed in the past the Smart City concept, which is the idea of utilizing an urban area with new electronic methods to provide services to citizens or collect data.

The newest solution from ArcRAN is committed to protecting software and hardware applications that are used in any smart facility or building and public areas. ArcRAN integrates its solution with IoT security detection APIs for tracking and scans local networks for unauthorized devices and behavior.

Their IoT Honeypot gains information and motives or tactics of attackers. Also, it can lure attackers into interacting with fake systems. 5G security is also utilized by the company both locally and privately.

Technologies relying more and more on 5G security could mean more cities adopting 5G to help assist the building of a true Smart City concept.

Blockchain Traceability

Numbers Protocol was one of 100 start-ups to be featured at CES 2021 from Taiwan. The company leverages blockchain to create The Numbers Capture app that can deal with debunking AI deep fakes and misinformation.

The app records birth information of a piece of media content and can generate a unique signature for that piece of content. Blockchain is then used to record the birth information, as well as any changes that happen to that piece of content in the future.

The Numbers Capture app also has a “PostCapture” function to create digital assets ensured by blockchain. So, for example, this function allows a way for an original image or asset to have reinstated value and security. The asset can still be shared and transferred to a new owner, but blockchain creates a level of security to prevent an asset from being tampered with.

This blockchain security technology can create a safer outlet for media outlets or even governments to release information and keep that information from being tampered with.

One-Click-Through AI Framework

IoT companies are always looking for smart implementations that can lead to faster optimization and operation of new technology.

Avalanche Computing announced at CES an AI framework that is meant to increase production speed and provide a smart implementation that requires one-click-through.

This means that Avalanche Computing’s AI framework can retrain another company’s AI model without having to change any algorithms. The framework would reduce AI project costs and deployment to other devices could occur in just one command.

From smart manufacturing to medical AI, to construction, Avalanche Computing hopes to not only increase manufacturing speed, but also add predictive maintenance, defect detection, and supply chain analysis.

New AI frameworks that are more interchangeable can lead businesses across industries to increase productivity, especially IoT companies, and save time on these types of projects in the future.

Emerging tech trends continue to play their role in revolutionizing business and enterprise in any industry and for any size of business. CES 2021 seemed to be no different from CES 2020 in continuing the ongoing trend in new technology that can assist with security, speed, and production.

Want to dive deeper into the best of emerging tech in 2021? Contact us at info@quantilus.com for a consultation and learn more about what Quantilus has to offer here.

The Best of CES 2021: 5 Product Innovations to Watch in 2021

January 14th, 2021|Categories: AI, NLP, Machine Learning, Emerging Tech - AR, Cloud, Blockchain|Tags: |

As the first-ever virtual CES 2021 comes to a close, let’s look back at, or forward to, some of the highlights of the unique version of consumer technology’s yearly showcase. From health and safety to the usual innovations in consumer tech, this all-digital CES still left viewers and attendees with tons of new products and concepts to talk about. Here are a few innovations geared towards navigating the changes occurring all over the world.

Best of CES 2021 in Health & Safety: Razer’s Project Hazel

A mask became a necessity in 2020 as COVID-19 shifted the way people needed to navigate their daily lives. Razer announced a new project at CES 2021 showcasing what they call the world’s smartest mask that boasts safety and social features, as well as being a non-disposable mask.

The mask features a surgical N95 respirator and has a high fluid resistance. The primary safety feature and differentiator for Project Hazel is the auto-sterilization function. The mask comes with a chargeable case that has a UV light interior that kills bacteria and potential viruses while the mask is charging.

From a social standpoint, the new mask features a transparent design, a low light mode that helps mask wearers still show their expressions in dark areas or environments, and has a built-in mic and amplifier. The most common complaints from mask wearers try to get addressed by Project Hazel. When people feel they’re not their usual selves when needing to wear a mask or don’t feel appropriately understood when speaking, Project Hazel is looking to up the comfort.

Speaking of comfort, the mask’s ventilation brings in cool air and doesn’t trap heat or CO2, and no part of the adjustable mask rests on the wearer’s mouth.

The expected release of the mask to consumers is by the end of 2021.

Best of CES 2021 in Productivity: Kensington’s StudioDock

Kensington’s StudioDock is exclusive for Apple users; however, it’s the future answer to the problem many Apple users have, especially those working from home, of trying to find the best set-up to utilize Apple products for work and actively charge them.

The StudioDock provides a compact place to charge your iPhone and AirPods, as well as a dock for most iPads currently on the market. The port can allow the iPad to act as a mini iMac, and the dock can wirelessly connect to an iMac so your iPad can become an extra screen for your home office. There is also optional charging support for an Apple Watch.

Kensington’s StudioDock ultimately offers flexibility to the user’s workspace better than some other competitors currently on the market. The dock also has a USB-C port, three USB-A ports, an audio jack, an HDMI port, an SD card reader, and an Ethernet jack. So for those not looking to invest in a new computer, the dock provides an option to turn your tablet into a mini-desktop computer.

There’s currently no release date set yet for the StudioDock.

Best of CES 2021 in Accessibility: CareClever’s Cutii

Senior care is one area that 2020, unfortunately, brought much change to and broke down support networks for many elders. CareClever debuted a robotic answer to help those feeling isolated with Cutii. Cutii is a companion robot and one of the first that is ready to be released in 2021.

Cutii’s differentiator is that it is entirely mobile, so no voice-activation is needed to use the companion. If, for example, an elder falls and needs assistance, Cutii will answer your call and try to get in touch with an emergency contact on its own. Like many other smart applications and devices, Cutii also can remind users of appointments.

Voice and video calls can also run through Cutii, and the companion can also act as a platform to host teleconsultations with doctors.

Cutii commercially launched in France, and CES 2021 kicked off the start-up’s marketing efforts in the United States.

Best of CES 2021 in Cleanliness: Roborock S7

With more time spent in the home, chances are the messes become more frequent as well. The Roborock S7 is a smart vacuum and mop that scrubs the floor up to 3,000 times per minute to clean the more challenging to get out messes it identifies. The mopping feature also identifies surfaces like rugs or carpets to ensure those surfaces don’t get wet.

The cleaning device’s corresponding app also gives users the ability to schedule when each room should be cleaned or if a particular part of the living space is a No-go Zone. Users can also adjust suction power at any time.

The Roborock S7 releases to consumers at the end of March and sells for $649.

Best of CES 2021 in Software: Owl Labs’ Whiteboard Enhance

The classroom or office definition has varied in the past year with changing circumstances due to the pandemic; however, Owl Labs’ Whiteboard Enhance provides a solution to help participants feel comfortable as hybrid meetings or classrooms continue to be a reality.

The Whiteboard Enhance software works in businesses and classrooms that already utilize Meeting Owl Pro and gives any user the ability to zoom in or out of any images on the whiteboard. For example, a student using the software can zoom in on the teacher’s board and not obstruct the view that other students have on their screens.

Meeting Owl Pro provides teachers or businesses to present in a 360-degree smart video conferencing environment.

CES 2021 saw innovations in areas that aren’t always a focal point at the yearly tech show; however, the best of CES 2021 still could bring excitement and maybe even some reassurance for those looking to make their lives a little more comfortable in 2021.

Contact us at info@quantilus.com for a consultation and learn more about what Quantilus has to offer here.

Gartner’s Top Nine Tech Trends of 2021

December 18th, 2020|Categories: AI, NLP, Machine Learning, Emerging Tech - AR, Cloud, Blockchain, News and Announcements|

A new year means new technology and advancements. With a seemingly endless amount of technology and innovations being created daily, it can be challenging for businesses to pinpoint which technologies are worth the investment. In 2020, the impact of the COVID-19 pandemic resulted in immense disruption for organizations, influencing technology adoption and requiring agile strategies. Global research and advisory firm, Gartner, recently released their predictions for the upcoming year’s top tech trends that CIOs and executives need to consider. Many of these trends highlight how organizations can maintain continuity amid changing or unforeseen conditions moving forward. Gartner’s tech trends for 2021 are as follows… 

  1. Internet of Behaviors (IoB)
  2. Total Experience
  3. Privacy-Enhancing computation
  4. Distributed cloud
  5. Anywhere Operations
  6. Cybersecurity mesh
  7. Intelligent composable business
  8. AI engineering
  9. Hyperautomation

Continue reading to explore each trend and find out their benefits. 

1) Internet of Behaviors:

The Internet of Behaviors, also referred to as IoB, utilizes collected user data to influence behaviors. This data is mined from various sources, including actions on a company’s website, social media profiles, computer vision, sensors or RFID tags, and more.  

An example of this is setting up computer vision on work devices. Computer vision can analyze employees’ faces and detect if they are wearing a face mask. If they are not wearing a face mask, it can remind the employee and alert this information to other stakeholders in the company.

Accordingly, there are ethical implications with the IoB as it collects data on users and attempts to change their behaviors. The data collected can also have varying levels of intrusiveness due to the technology being able to capture a wide variety of personal information. Laws on the IoB vary widely, which will play a critical role in adopting and advancing this tech trend.

      

2) Total Experience:

Total experience links the customer, user, and employee experiences together using multi-experience. Multi-experience utilizes various technologies to engage and interact with consumers. Therefore, total experience focuses on leveraging different technologies to create a more seamless process for customers, users, and employees. 

Gartner’s tech trends report provided an example of a telecommunications company improving its total experience by interlinking both customer and employee in-store experiences to enforce social distancing guidelines. Specifically, the telecommunications company leveraged its consumer-facing app for scheduling appointments, geo-tracking, and push notifications to manage the customer’s in-store experience in a safe, orderly way. Simultaneously, employees in the store leveraged digital kiosks to service the customer as well as personal tablets to co-browse customers’ devices, thus avoiding the need to touch consumers’ devices.

So instead of perfecting the customer, user, and employee experiences independently, they should be thought of as one experience and flow harmoniously into each other. Meanwhile, integrating various technology solutions into these experiences will help automate tasks, improve efficiencies, guide customers and staff, maintain social distancing guidelines, engage customers, and more. 

      

3) Privacy-enhancing computation

Privacy-enhancing computation aims to protect personal information in the following ways:

  • Provide a safe environment for sensitive data to be processed and analyzed. 
  • Perform processing and analytics in a decentralized manner. 
  • Transform data and algorithms before processing or analytics.

Privacy enhancing computation is becoming increasingly important because as teams become more geographically widespread and more people require access to sensitive information, the demand for digital and online protection also grows. Privacy enhancing computation also allows for better collaboration while not compromising security. 

 

4) Distributed cloud

The purpose of distributed cloud is to have cloud providers implement their services near their customers. Perks of having cloud services physically closer to customers include reducing low-latency, decreasing data costs, and complying with laws that require data to remain in a specified geographical location. Additionally, businesses are not accountable for managing their cloud when using distributed cloud, which can be an expensive and elaborate responsibility.

 

5) Anywhere Operations

Anywhere operations is an IT model that allows businesses to operate…from anywhere. This means employees can work remotely and companies can operate from nearly anywhere in the world, and still complete their tasks. However, this does not mean that employees and businesses must always be remote. 

 Anywhere operations require businesses to become digital-centric. This means digital processes and operations should be the default for organizations. Therefore, business is less likely to be interrupted due to things such as social distancing guidelines and inclement weather. Lastly, physical stores still have a place in business, but the stores should adopt digital processes like contactless payment. Anywhere operations minimizes downturns in business due to foreseen and unforeseen circumstances (such as COVID-19).

 

6) Cybersecurity mesh

An increase in remote workers equates to more opportunities for online security to be compromised. Cybersecurity mesh allows users to safely access digital information, files, and assets securely from anywhere geographically.  

To create a cybersecurity mesh, companies need to take security measures outside of the ‘normal’, physical location where information was originally intended to be accessed…which in most cases is the company office(s). To do so, businesses should implement scalable cloud models that set user identity as the security parameter. This way, employees can securely access company information and assets anywhere around the globe without worrying about data leaks. 

 

7) Intelligent Composable Business

Successful businesses can quickly adapt their processes, operations, and model based on data. Therefore, intelligent composable business means as technology continues to transform, it is vital that companies:

  • Have greater access to all pertinent information.
  • Gather more insightful findings from the information.
  • Adapt their model based on their findings.

Furthermore, autonomy will be essential in helping companies quickly adapt so outdated processes do not delay businesses. Find out more information on automation by reading trend nine below. 

 

8) AI Engineering

Gartner’s findings reveal that only 53% of artificial intelligence projects make it from prototype to production. Therefore, many AI projects appear to lack resources, scalability, and internal company support. 

That’s where AI engineering comes into the equation. The foundation of AI engineering is made of three pillars – DataOps, ModelOps, and DevOps. Thus, AI engineering is essentially lifecycle management for artificial intelligence projects. More planning, strategy, team-members, financial backing, etc., increases the likelihood of AI projects making it to production. Therefore, giving the AI projects a chance to make a profit. Also, it’s important to note that once an AI project does come to production, it is easier for the product to evolve as new technology and findings are revealed. 

 

9) Hyperautomation

As Gartner’s top tech trend of 2020, hyperautomation landed another spot on this year’s list. Hyperautomation is automating anything in a business that can be automated. If processes are left un-automated, companies are likely operating less efficiently than they could be and are therefore missing out on profit. Working at less than optimal efficiency also makes organizations more susceptible to falling behind competitors. 

 

Conclusion

Businesses in all industries can implement these tech trends, not just companies in the technology sector. Furthermore, companies can cherry-pick which trends they wish to implement in their business. However, it is essential to note that technology tends to be more efficient and successful when combined with other technologies. 

Which tech trends do you think will be the most vital during 2021? Let us know in the comments below. Interested in learning more about how technology can help your business? Contact us at info@quantilus.com for a consultation to learn how we can help.

Bitnami