Newsletter

Newsletter
 

Media & Computer Vision


Media organizations, internet companies, and technology development enterprises are finding more and more needs for technical computing in research and development departments. The pervasiveness of GPUs across enterprises and the need for better, faster, and cheaper ways of performing modeling, simulation, and analytics make ArrayFire a perfect solution for industries looking for high performance applications.

Companies such as Google and Adobe have found that ArrayFire provides a unique way to leverage the power and flexibility of GPU computing without the heavy investment in development time and resources.


Example Applications

ArrayFire can help educate students and support research in many application areas including:

  • Knowledge Discovery on Sparse Data
  • Video processing and Analytics
  • Mathematics
  • Statistics
  • Image analysis
  • Visualization
  • And much more...

Fast Computer Vision with OpenCV and ArrayFire
OpenCV Blogger
Speedup: 10X


OpenCV image

Fast Computer Vision with OpenCV and ArrayFire

Authors: OpenCV Blogger
Speedup: ~10X

The OpenCV library is the defacto standard for doing computer vision and image processing research projects. OpenCV includes several hundreds of computer vision algorithms, aimed for use in realtime vision applications. This case study shows how to use both libraries together. There is a simple example application that demonstrates using OpenCV for webcam access and ArrayFire for some basic processing routines and displaying results.

 

Last Updated: 24 Aug 2011

Video Processing
Google
Speedup: 10X - 20X


Google Video Processing image

Video Processing

Authors: Google and Stanford University
Speedup: 10 to 20X

Video content analysis is the basis for categorizing videos and enabling search by content. Growing interest in using sparse-coding methods to extract motion features in video in support of video content analysis led to the application of AccelerEyes software and GPUs to improve performance by substantially accelerating the solution of the L1-regularized least-squares optimization problem.

 

Last Updated: 13 Jan 2010

Action Recognition with Independent Subspace Analysis
Stanford University
Speedup: 4.4X


Feature Learning

Action Recognition with Independent Subspace Analysis

Authors: Quoc Le, Will Zou, Serena Yeung, Andrew Ng, Stanford University
Speedup: 4.4X

In a paper at this year's CVPR 2011, entitled "Learning hierarchical invariant spatio-temporal features for action recognition with independent subspace analysis", the authors explain how their unsupervised feature learning algorithm competes with other algorithms that are hand crafted or use learned features. For their training purposes, they used a multi-layered stacked convolutional ISA (Independent subspace analysis) network. An ISA is used for learning features from image patches without supervision.

Last Updated: 19 Aug 2011

Music Beat Analysis
Georgia Tech
Speedup: 15X


Music Beat Analyzer

Music Beat Analysis

Authors: Vidhur Vohra - Georgia Tech
Speedup: 15X

Did you ever wonder how the music visualizer in your media player works? Watching it pulsate in synchrony with the beats of the song is almost as entertaining as listening to the song itself! Researchers have been attempting to detect beats in audio signals for many years, and there are many techniques available, from the simplest (and least accurate) to more complicated algorithms that are highly accurate. All algorithms, though, perform some form of signal processing and frequency analysis, applications highly suited to GPU Computing.

Last Updated: 11 Aug 2011

Optimization methods for deep learning
Stanford Artificial Intelligence Laboratory
Speedup: Improved Accuracy


SAIL image

Optimization methods for deep learning

Authors: Stanford Artificial Intelligence Laboratory
Speedup: Improved Accuracy

Researchers at SAIL (Stanford Artificial Intelligence Laboratory), have done it again. They have successfully used AccelerEyes software to speed up the training part of Deep Learning algorithms. In their paper titled .On Optimization Methods for Deep Learning., they experiment with some of the well known training algorithms and demostrate their scalability across parallel architectures (GPUs as well as multi-machine networks). The algorithms include SGDs (Stochastic Gradient Descent) L-BFGS (Limited BFGS used for solving non-linear problems), CG (Conjugate Gradient).

 

Last Updated: 20 Sep 2011

Feature Learning on Images
Stanford University
Speedup: Hours of runtime reduction


Feature Learning

Feature Learning Architectures with GPU-acceleration

Authors: Andrew Ng, Stanford University
Speedup: Ability to process many images in parallel

Stanford researchers in Andrew Ng’s group used AccelerEyes software to speed up their work on Feature Learning Architectures. They decided to use GPUs and AccelerEyes software for this study because of the need to quickly evaluate many architectures on thousands of images.

Last Updated: 9 Apr 2011

Digital Holography for Imaging
National University of Ireland, Maynooth
Speedup: 17X


Digital Holography

Digital Holography

Authors: Nitesh Pandey, Damien Kelly, Bryan Hennelly and Thomas Naughton from the National University of Ireland, Maynooth
Speedup:17X

Digital holography is a powerful imaging technique with many new applications like true 3D display. It allows the capture of both amplitude and phase information of the light reflected off the surface of 3D objects. Researchers at the National University of Ireland, Maynooth are developing techniques based on digital holography for 3D display applications.
Reconstruction of large digital holograms can be computationally intensive to generate on CPUs, but GPUs running AccelerEyes software offer amazing possibilities.

Last Updated: 30 Apr 2011