Coming later this spring, the Kinect for Windows SDK is a programming toolkit that will enable researchers and enthusiasts easy access to the capabilities offered by the Microsoft Kinect device connected to computers running Microsoft Windows 7.
Microsoft Tag connects real life with the digital world. Microsoft Tags are small, colorful codes that can be printed, stuck, or displayed just about anywhere. When you snap a Tag with the camera on your internet-enabled phone, additional information or experiences are automatically opened on your phone. There is no fumbling with URLs or texting short codes. Microsoft Tags can make product packages, posters, print-based ads, magazine articles, exhibit signage, billboards, storefronts, business card, or just about anything else, interactive.
Songsmith generates musical accompaniment to match a singer’s voice. Just choose a musical style, sing into your PC’s microphone, and Songsmith will create backing music for you. Then share your songs with your friends and family, post your songs online, or create your own music videos.
The WorldWide Telescope (WWT) is a rich visualization environment that functions as a virtual telescope, bringing together imagery from the best ground- and space-based telescopes in the world to enable seamless, guided explorations of the universe.
Personal Web Site
- Microsoft Research Redmond
One Microsoft Way
Redmond, WA 98052 U.S.
- Avoiding Vulnerable Passwords—and Rules, Too
Microsoft Research has released a free tool to help you avoid choosing bad passwords
Thu, 05 Dec 2013 17:00:00 GMT
- Studio 99 Melds Art and Science
For decades, there have been artist-in-residency programs within scientific and academic communities. An exhibit in Redmond, Wash., presents the results of Microsoft Research’s collaborations with its first artist-in-residence, James George.
Tue, 03 Dec 2013 18:00:00 GMT
- A New Tool for Teaching Climate Change in the Cloud
ChronoZoom, an open-source-community project from Microsoft Research, makes it easy to collect, present, and manage a huge amount of educational resources.
Wed, 20 Nov 2013 18:00:00 GMT
- Grudin’s Work Cited for ‘Lasting Impact’
Jonathan Grudin of Microsoft Research has been honored with the inaugural CSCW Lasting Impact Award for a paper he wrote in 1988. The award will be presented in Baltimore in February during the CSCW conference.
Tue, 19 Nov 2013 21:40:00 GMT
- Learn How Microsoft Researchers Are Using Wearable Technology to Read Your Moods
Mary Czerwinski and Asta Roseway of Microsoft Research are investigating “affective computing,” systems that try to identify your current mood and react accordingly.
Tue, 19 Nov 2013 13:00:00 GMT
- Automated video looping with progressive dynamism
An application to extract a 5 second looping video from a non-looping input video. Also a viewer application to enable both interactive control over the level of dynamism of the output video, as well as manual editing of which regions animate or are static.
Wed, 04 Dec 2013 00:20:07 GMT
- Local Deep Kernel Learning
There has been an explosion in the size of modern day training sets with the advent of big data, cheap crowdsourcing and other techniques for gathering labelled information efficiently. This presents a significant challenge for non-linear SVMs since their cost of prediction can grow linearly with the size of the training set. Thus, even though non-linear SVMs have defined the state-of-the-art on multiple benchmark tasks, their use in real world applications remains limited. We develop a Local Deep Kernel Learning (LDKL) technique for efficient non-linear SVM prediction while maintaining classification accuracy above an acceptable threshold. LDKL learns a tree-based primal feature embedding which is high dimensional and sparse. Primal based classification decouples prediction costs from the number of support vectors and the size of the training set and LDKL’s tree-structured features efficiently encode non-linearities while speeding up prediction exponentially over the state-of-the-art. We develop routines for optimizing over the space of tree-structured features and efficiently scale to problems with more than half a million training points. Experiments on benchmark data sets reveal that LDKL can reduce prediction costs by more than three orders of magnitude in some cases with a moderate sacrifice in classification accuracy as compared to RBF-SVMs. Furthermore, LDKL can achieve better classification accuracies over leading methods for speeding up non-linear SVM prediction. In particular, LDKL is significantly better than kernel approximation techniques, such as Random Fourier Features and Nyström, as it focusses on the decision boundary rather than modeling the kernel everywhere in space. LDKL can also be much faster than reduced set methods as its tree-structured features can be computed in logarithmic time.
Wed, 27 Nov 2013 01:11:19 GMT
- Mood Board 1.3
Microsoft Research Mood Board is a Windows application that combines image search, image collection, and sketching to support creative activities.
Thu, 21 Nov 2013 03:04:53 GMT
- Contig ploidy and allele dosage estimation (ConPADE)
ConPADE is a tool used to estimate contig ploidy and allele dosage in polyploid genome assemblies.
Mon, 18 Nov 2013 21:13:46 GMT
- FaST-LMM - Factored Spectrally Transformed Linear Mixed Model
FaST-LMM (Factored Spectrally Transformed Linear Mixed Models) is a program for performing genome-wide association studies (GWAS) on large data sets. It runs on both Windows and Linux system, and has been tested on data sets with over 120,000 individuals.
Wed, 13 Nov 2013 07:11:27 GMT