The Apollo 11 lunar landing may have been “one small step” for Armstrong and crew and a “giant leap for mankind,” but recently, it’s been a pretty significant event for graphics giant NVIDIA.
Armed with NVIDIA’s Tesla GPUs and CUDA programming technology, video restoration provider Lowry Digital was able to restore the lost July 20, 1969 television footage of that historic event and enhance it so it could be revisited and remembered in high-definition (HD) fidelity. The Tesla GPU/CUDA duo provided high-performance parallel processing horsepower to develop and use complex image processing algorithms that enabled the restoration.
Tesla GPU work together with CPUs, in a co-processing mode, allowing computationally-intensive applications like digital video restoration to be deployed to the high-power GPU while the sequential part of the application’s code runs on traditional CPU. The result: a 100-times boost in performance, Digital Lowry officials say. For example, enhancing each video frame on a CPU-only system would have taken between 20 to 45 minutes to complete; the Tesla GPU set up cuts back that time for a single frame to seconds.
To piece together the footage, Lowry Digital tapped several video sources, including low-quality images such as television broadcast video and 8mm film shot on a handheld camera that was pointed at the monitor at NASA’s Honeysuckle Creek tracking station in Australia. With the system, Lowry Digital was able to remove artifacts like noise, flickering, darkened image corners, blurs and smears-the result being improved contrast and resolution. The final Apollo 11 video will feature two and half hours of HD video.
Are they robots or androids? We're not exactly sure. Each talking, gesturing Geminoid looks exactly like a real individual, starting with their creator, professor Hiroshi Ishiguro of Osaka University in Japan.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.