Adobe Creative Cloud

December 12, 2016 /Digital Imaging /

How Facial Recognition and Detection Technology Will Influence Our Creative Future

Facial detection and recognition has evolved from a futuristic fairytale into a central piece of today’s digital, social, and technological landscape. Consider Facebook’s facial recognition software. Users upload a photo to Facebook and their facial recognition algorithm goes to work instantly, taking stock of the people in the image and tagging them. What’s the distance between her eyes and nose? Does this face align with faces in existing Facebook photos? The process is eerily accurate – and they’re just getting started.

Facebook is not alone. Snapchat’s filters – one of the social platform’s hottest features, which dramatically increases users’ ability for creative self-expression – use advanced facial detection. According to recent analysis from VOX, Snapchat’s intelligent detection tools rely on data from thousands of faces to identify individual facial features, create a point-mask that matches up to the data being received from a camera at 24 frames per second, and then develops a mesh mask which moves with the user’s face in real-time. Facial recognition isn’t just for processing tools and social networks, either. Cameras incorporate a version of this technology which helps keep images from blurring. Facial recognition and detection is changing the creative process at both the moment of creation and during post-processing, providing tighter creative controls and making it easier to achieve bold artistic visions.

These tools represent the evolution of years of R&D in the advanced imaging space. Two streams of imaging research are coming together. On one end, 3D imaging is providing richer representations of image content: dimensions, materials, and even behaviors such as the ability to animate. At the same time, sophisticated research is driving dramatic advances in computer vision, machine learning, computational photography, and more. Together, both streams of research are helping us reach the goal of developing the sophisticated underpinnings needed to accurately identify what is in an image. In turn, this allows software tools to automate more of the processing and lets artists focus on the visions they’re creating.

Facial detection technology in Photoshop Fix.

Facial detection technology in Photoshop Fix

Adobe’s own Face-Aware Liquify feature in Photoshop Fix and Photoshop CC has quickly embedded itself into creatives’ toolkits. From professional retouchers making slight adjustments working in the fashion industry to caricature artists taking a more drastic approach to manipulating facial features, Face-Aware Liquify has shown how much versatility one tool can offer creatives across industries.

Face-Aware Liquify in Photoshop CC

Face-Aware Liquify in Photoshop CC

 

Face-Aware Liquify

Face-Aware Liquify

And the technology is just scratching the surface of what advanced imaging technology will allow artists, photographers and filmmakers to do over the next few years. In many ways, these technological innovations are driven by what artists are already doing; we’re simply finding ways to streamline the process. To quote John Lasseter, “Art challenges technology, and technology inspires art.” The two exist in a synergistic system with each one driving the other and both co-evolving together. What we are seeing now – both with facial recognition and detection technology and how the creative community is embracing them – is just the tip of the iceberg.

What does this mean for the creative process and for creatives who are working to keep up with today’s high demand for content? A lot less complication. If you are a creative tasked with putting together an advertisement for a car, for example, you don’t need to determine or deal with the logistics of a complex shoot involving shutting down roads, setting up lighting, and driving a car numerous times through your environment. Instead, you have a 3D model which can be composited and placed into just about any environment you can imagine. This allows the artist to focus on all the elegant ways to showcase the car, instead of worrying about the mechanics involved in doing so. Technology, in effect, empowers artists to rethink how they approach their art and frees them up to focus more on creative direction, rather than logistical minutiae.

While this technology is empowering artists, they’re having a similar effect on what we as software developers decide to create for them. In many ways creatives have already devised workarounds using existing tools and are starving for new, automated solutions that address their needs without a complex learning curve. Something that they can quickly and easily integrate into their existing creative process. So how can creatives prepare for what’s to come? You can both prepare and help shape the future by continuing to push the boundaries with existing tools, expanding your personal arsenal by experimenting with 3D, VR, 360-degree video, photography and other immersive experiences, and continuing to challenge technology to keep up with your creative vision.

With the convergence of so many technologies driving new features and functionality, it begs the question, “What’s next in the world of digital imaging?”

Computer vision is one of the holy grails of imaging. When the point is reached where the algorithms automatically understand what is in an image, there will be a profound effect on how users can interact with them. A system which can differentiate between metal, fabric and skin empowers artists in new and unique ways. Machine learning has been used to match 3D models into images and videos for several years, but now the quality is finally getting very close to being production level. Hence the inflection point.

Our goal in this field, then, is simple: empower creatives from all industries, backgrounds and mediums to focus on their art by automating more of the drudgery that until now has come with the process of creating an end product. The more sophisticated we get with 3D techniques and the better we get with matching 3D to images, the more we can empower creative users and facilitate them in realizing their unique artistic visions. Overall, though, it is about less mechanics and more vision, a better balance for any artist, no matter their approach, their medium, or their desired creative output – and that is an exciting prospect for all creatives.

Digital Imaging