Fast, Accurate Detection of 100,000 Object Classes on a Single Machine a prizewinning paper by Google Research scientists, describes a breakthrough in machine vision that can distinguish between a huge class of objects 20,000 times faster than before.
This so-called convolution operator is one of the key operations used in computer vision and, more broadly, all of signal processing. Unfortunately, it is computationally expensive and hence researchers use it sparingly or employ exotic SIMD hardware like GPUs and FPGAs to mitigate the computational cost. We turn things on their head by showing how one can use fast table lookup — a method called hashing — to trade time for space, replacing the computationally-expensive inner loop of the convolution operator — a sequence of multiplications and additions — required for performing millions of convolutions with a single table lookup.
We demonstrate the advantages of our approach by scaling object detection from the current state of the art involving several hundred or at most a few thousand of object categories to 100,000 categories requiring what would amount to more than a million convolutions. Moreover, our demonstration was carried out on a single commodity computer requiring only a few seconds for each image. The basic technology is used in several pieces of Google infrastructure and can be applied to problems outside of computer vision such as auditory signal processing.
Fast, Accurate Detection of 100,000 Object Classes on a Single Machine
(Image: Clutter, a Creative Commons Attribution Share-Alike (2.0) image from neofob's photostream)
Last June, an independent tribunal concluded that the Chinese state was nonconsensually harvesting organs from prisoners despite promises that the practice had ended in 2014.
Princeton's Center for Information Technology Policy is a marvellous interdisciplinary research center, and it is advertising for "visitors" for one-year stints: postdocs, policy fellows and visiting IT professors.
Geneva ("Genetic Evasion") is a project from the University of Maryland's Breakerspace ("a lab dedicated to scaling-up undergraduate research in computer and network security"); in a paper presented today at the ACM's Conference on Computer and Communications Security, a trio of Maryland researchers and a UC Berkeley colleague present their work on evolutionary algorithms as […]
Need a boost on that resume? Get a valuable tech education on your own time with these eBook bundles. They contain guides from Packt Publishing that cover everything from game development to machine learning. The Complete Mobile App Developer eBook Bundle It’s a veritable gold rush in the App Store these days. Get in on […]
Vinyl is officially back. People are hearing the proof behind the initial “retro” excitement: that records really do have a richer sound. And if you haven’t switched to old-school records for serious listening, it’s a new golden age. Why? Because quality turntables like the Altec Lansing ALT-500 are finally available to a market other than […]
Between all of our apps, streaming devices, Bluetooth speakers, and energy-sucking decorations, paying for utilities each month can be…brutal. In fact, the average household spends roughly $70 a month on the water bill alone. That number might not seem terribly significant, but when you add it up, that’s $840 a year — a pretty significant […]