Stories, Papers, WIKIs

Title Body
Optimization of Tele-Immersion Codes

Abstract:

As computational power increases, tele-immersive applications are an emerging trend. These applications make extensive demands on computational resources through their heavy use of real-time 3D reconstruction algorithms. Since computer vision developers do not necessarily have parallel programming expertise, it is important to give them the tools and capabilities to naturally express computer vision algorithms, yet retain high efficiency by exploiting modern GPU and large-scale multi-core platforms. In this paper, we describe our optimization efforts for a tele-immersion application by tuning it for GPU and multicore platforms. Additionally, we introduce a method to obtain portability, high performance, and increase programmer productivity.

Slides from GPGPU-2

Feature Tracking and Matching in Video Using Programmable Graphics Hardware

Abstract:

 

This paper describes novel implementations of the KLT feature-tracking and SIFT feature extraction algorithms that run on the graphics processing unit and are suitable for video analysis in real-time vision systems. While significant acceleration over standard CPU implementations is obtained by exploiting parallelism provided by modern programmable graphics hardware, the CPU is freed up to run other computations in parallel.

 Our GPU- based KLT implementation tracks about a thousand features in real-time at 30 Hz on 1024 x 768 resolution video which is a 20 times improvement over the CPU. The GPU-based SIFT implementation extracts about 800 features from 640 x 480 video at 10 Hz which is approximately 10 times faster than an optimized CPU implementation.
Continuous Maximal Flows and Wulff Shapes: Application to MRFs

 

Abstract:
 
 Convex and continuous energy formulations for low level vision problems enable efficient search procedures for the corresponding globally optimal solutions. In this work we extend the well-established continuous, isotropic capacity-based maximal flow framework to the anisotropic setting. By using powerful results from convex analysis, a very simple and efficient minimization procedure is derived.

 Further, we show that many important properties carry over to the new anisotropic framework, e.g. globally optimal binary results can be achieved simply by thresholding the continuous solution. In addition, we unify the anisotropic continuous maximal flow approach with a recently proposed convex and continuous formulation for Markov random fields, thereby allowing more general smoothness priors to be incorporated. Dense stereo results are included to illustrate the capabilities of the proposed approach....The underlying update equations Eq. 12 are very suitable to be accelerated by a modern GPU. Our current CUDA-based implementation executed on a Geforce 8800 Ultra is able to achieve two frames per second for 320 × 240 images and 32 disparity levels (aiming for a 2% duality gap at maximum).

 

 
From Structure-from-Motion Point Clouds to Fast Location Recognition

Abstract:

 

Efficient view registration with respect to a given 3D reconstruction has many applications like inside-out tracking in indoor and outdoor environments, and geo-locating images from large photo collections. We present a fast location recognition technique based on structure from motion point clouds.

 Vocabulary tree-based indexing of features directly returns relevant fragments of 3D models instead of documents from the images database. Additionally, we propose a compressed 3D scene representation which improves recognition rates while simultaneously reducing the computation time and the memory consumption. The design of our method is based on algorithms that efficiently utilize modern graphics processing units to deliver real-time performance for view registration. We demonstrate the approach by matching hand-held outdoor videos to known 3D urban models, and by registering images from online photo collections to the corresponding landmarks. 

 

...we employ a CUDA-based approach executed on the GPU for faster determination of the respective visual words. The speed-up induced by the GPU (about 15 - 20 on a GeForce GTX280 vs. Intel Pentium D 3.2Ghz) approach allows to incorporate more descriptor comparisons, i.e. a deeper tree with a smaller branching factor can be replaced by a shallower tree with a significantly higher number of branches.
Bayesian real-time perception algorithms on GPU

Abstract:  

 

In this text we present the real-time implementation of a Bayesian framework for robotic multisensory perception on a graphics processing unit (GPU) using the Compute Unified Device Architecture (CUDA). As an additional objective, we intend to show the benefits of parallel computing for similar problems (i.e. probabilisticgrid-based frameworks), and the user-friendly nature of CUDA as a programming tool. Inspired by the study of biological systems, several Bayesian inference algorithms for artificial perception have been proposed. Their high computational cost has been a prohibitory factor for realtime implementations. However in some cases the bottleneck is in the large data structures involved, rather than the Bayesian inference per se. We will demonstrate that the SIMD (single-instruction, multiple-data) features of GPUs provide a means for taking a complicated framework of relatively simple and highly parallelisable algorithms operating on large data structures, which might take up to several minutes of execution with a regular CPU implementation, and arrive at an implementation that executes in the order of tenths of a second. The implemented multimodal perception module (including stereovision, binaural sensing and inertial sensing) builds an egocentric representation of occupancy and local motion, the Bayesian Volumetric Map (BVM), based on which gaze shift decisions are made to perform active exploration and reduce the entropy of the BVM. Experimental results show that the real-time implementation successfully drives the robotic system to explore areas of the environment mapped with high uncertainty. 
Applying Parallel Design Techniques to Template Matching with GPUs

   Designing algorithms to be parallel from the beginning can lead to substantial performance increases. In this paper we present an algorithm for template matching that is targeted from the beginning for the GPU architecture and achieves greater than an order of magnitude speedup over traditional algorithms designed for the CPU and reimplemented on the GPU. This shows that it is not only desirable to adapt existing algorithms to run on GPUs, but also that future algorithms should be designed with a multi-core architecture in mind. Its speedup factor is nearly linear in p, and its overall cost is not far from serial implementation. 
Accelerating Viola-Jones Face Detection to FPGA-Level using GPUs

   Face detection is an important aspect for biometrics, video surveillance and human computer interaction. We present a multi-GPU implementation of the Viola-Jones face detection algorithm that meets the performance of the fastest known FPGA implementation. The GPU design offers far lower development costs, but the FPGA implementation consumes less power. We discuss the performance programming required to realize our design, and describe future research directions. 
OpenVIDIA: Parallel GPU Computer Vision

Abstract:

 

Graphics and vision are approximate inverses of each other: ordinarily Graphics Processing Units (GPUs) are used to convert \numbers into pictures" (i.e. computer graphics). In this paper, we propose using GPUs in approximately the reverse way: to assist in converting pictures into numbers" (i.e. computer vision). The OpenVIDIA project uses single or multiple graphics cards to accelerate image analysis and computer vision. It is a library and API aimed at providing a graphics hardware accelerated processing framework for image processing and computer vision. OpenVIDIA explores the creation of a parallel computer architecture consisting of multiple Graphics Processing Units (GPUs) built entirely from commodity hardware. OpenVIDIA uses multiple Graphics Processing Units in parallel to operate as a general-purpose parallel computer architecture. It provides a simple API which implements some common computer vision algorithms. Many components can be used immediately and because the project is Open Source, the code is intended to serve as templates and examples for how similar algorithms are mapped onto graphics hardware. Implemented are image processing techniques (Canny edge detection, ltering), image feature handling (identifying and matching features) and image registration, to name a few.
Using Graphics Devices in Reverse: GPU-based Image Processing and Computer Vision

Abstract:

 

Graphics and vision are approximate inverses of each other: ordinarily Graphics Processing Units (GPUs) are used to convert “numbers into pictures” (i.e. computer graphics). In this paper, we discus the use of GPUs in approximately the reverse way: to assist in “converting pictures into numbers” (i.e. computer vision). For graphical operations, GPUs currently provide many hundreds of gigaflops of processing power. This paper discusses how this processing power is being harnessed for Image Processing and Computer Vision, thereby providing dramatic speedups on commodity, readily available graphics hardware. A brief review of algorithms mapped to the GPU by using the graphics API for vision is presented. The recent NVIDIA CUDA programming model is then introduced as a way of expressing programparallelism without the need for graphics expertise.
GpuCV: A GPU-accelerated framework for Image Processing and Computer Vision

This paper presents briefly describes the state of the art of accelerating image processing with graphics hardware (GPU) and discusses some of its caveats. Then it describes GpuCV, an open source multi-platform library for GPU-accelerated image processing and Computer Vision operators and applications. It is meant for computer vision scientist not familiar with GPU technologies. GpuCV is designed to be compatible with the popular OpenCV library by offering GPU-accelerated operators that can be integrated into native OpenCV applications. The GpuCV framework transparently manages hardware capabilities, data synchronization, activation of low level GLSL and CUDA programs, on-the-fly benchmarking and switching to the most efficient implementation and finally off ers a set of image processing operators with GPU acceleration available.