Intel has added yet another important name to its graphics research efforts. Anton Kaplanyan, a Ph.D. in light transport who played an important role in NVIDIA's RTX and DLSS developments, has joined Intel as Vice President of Graphics Research of Intel's AXG Group. Given his pedigree, the blue giant is likely looking for him to help develop the company's own super sampling solution. This rings especially true when one considers his July 2020 paper on Neural Supersampling, which does pave the way for related contributions to Intel.
Anton Kaplanyan invented the one-sample-per-pixel denoising approach used by NVIDIA in its DLSS implementation and was among the first to integrate neural networks and graphics technologies. His previous work includes a stint at game developer Crytek (who have had one of the most impressive game engines on their hands at one point in time in the form of CryEngine); co-founded Lightrig, a company specializing in digital illumination and control; and then followed up with his undertaking of RTX and DLSS on Nvidia. He finally settled with Facebook, directly working with the company's VR division in developing virtual reality graphics technologies.
Machine learning and light simulation have been the core technologies at the heart of Anton's contributions, and he does boast of an impressive resumé in that particular field. He explains it on his website, where he defines his work as: "I enjoy advancing research in computer graphics with a current focus on real-time rendering, including neural rendering, perceptual rendering, shading and appearance, as well as differentiable rendering. My research is used in Nvidia's middleware and RTX hardware, game engines (UE4, Unity, CryEngine) and games, and in Pixar's RenderMan. My Ph.D. is on light transport simulation, also a former game dev at Crytek.)."
Intel has been heavily investing in its graphics technologies ever since the "original" poaching of AMD's then-Senior Vice President and Chief Architect of Radeon Technologies Group Raja Koduri. The blog post penned by Anton on Intel's website represents an interesting, corporate-approved look at his future work in the company; as he puts it, Intel's resources "(...) gives us an opportunity to not only completely reinvent how we co-design graphics hardware and software with these technologies in the driver’s seat, but also to look at a larger picture of a potentially novel graphics ecosystem that can include heterogeneous distributed systems and multi-level compute for movie-quality visuals on a wide range of consumer devices."
Of course, the objective has always been lifelike visuals directly beamed into our retinas - and Anton may be a key piece in Intel's efforts to capitalize on that vision.