Baidu Launches Augmented Reality Platform for Smartphones

News, Research, Augmented Reality, Computer Vision & Machine Vision, CUDA, Image Recognition, Machine Learning & Artificial Intelligence, Media & Entertainment, Quadro

Nadeem Mohammad, posted Aug 04 2016

Baidu’s new DuSee platform allows people to make use of augmented reality within the Chinese internet giant’s apps, such as Mobile Baidu search, and takes advertising to the next level. In a demo of the technology, when a user of the Mobile Baidu app points their smartphone at a map of Shanghai, a virtual 3D

Read more

Share Your Science: Exploring the Wonders of the World in Virtual Reality

News, Research, CUDA, GeForce, Media & Entertainment, Quadro, Share Your Science, Virtual Reality

Nadeem Mohammad, posted Aug 03 2016

Dominic Eskofier, co-founder of shares how they are using GPUs for their photogrammetric virtual reality app, Realities on Steam. The app transports users to some of the most mesmerizing places around the world – like ancient Mayan temples in Guatemala – with the push of a button. Using CUDA with NVIDIA GeForce GTX 980

Read more

Video Technology Helps Pokémon Interact with Real World

Research, Augmented Reality, GeForce, Higher Education / Academia, Media & Entertainment

Nadeem Mohammad, posted Aug 02 2016

The augmented reality game Pokémon Go is by far the most popular mobile game ever created – but what if its virtual characters could interact with the real world? Researchers at MIT recently published a paper introducing a technique called “interactive dynamic video,” which lets people reach in and touch objects in videos. Using an

Read more

Reading Between The Threads: Shader Intrinsics

GameWorks, GameWorks Expert Developer, DX11, DX12, Vulkan, OpenGL, nvapi, DesignWorks

Mathias Schott, posted Jul 29 2016


When writing compute shaders, it’s often necessary to communicate values between threads. This is typically done via shared memory. Kepler GPUs introduced “shuffle” intrinsics, which allow threads of a warp to directly read each other's registers avoiding memory access and synchronization. Shared memory is relatively fast but instructions that operate without using memory of any kind are significantly faster still.

Read more