Projects

Approaches to Implement University of Washington
Mediaprocessor Computing Library (UW-MPCL)
Image Computing Systems Laboratory

Projects

Introduction to the University of Washington Mediaprocessor User Consortium (UW-MPUC)

Wed, 2006-11-22 15:07 — sluo

Image Computing Systems Laboratory
University of Washington
Seattle, Washington 98195 U.S.A.

1.0 Summary

University of Washington Mediaprocessor User Consortium (UW-MPUC) is an organization that aims to facilitate and promote the widespread use of programmable mediaprocessors (MPs). UW-MPUC will provide publications, internal reports, web site, training courses, software libraries, and objective evaluations of mediaprocessors, to increase the awareness of the benefits of mediaprocessors over application specific integrated circuits (ASICs), make the transition to mediaprocessor-based products easier for set makers, and help them attain full advantage of using mediaprocessors. UW-MPUC will also promote the development of standards to increase the portability and interoperability of mediaprocessor software. Member companies of UW-MPUC pay annual sponsorship fees to cover the expense of the consortium operations (e.g., library function development, training, publications, and other R&D activities). UW-MPUC will begin its operation in Fall 2000. Figure 1 shows the role of UW-MPUC. Governance of UW-MPUC will be via Consortium Advisory Board consisting of the entire UW-MPUC member companies in consultation with the University of Washington.

Projects

Mediaprocessor User Consortium

Wed, 2006-11-22 14:51 — sluo

Welcome to UW-MPUC!

University of Washington Mediaprocessor User Consortium.

Projects

Overview

Wed, 2006-11-22 14:50 — sluo

ICSL has been pursuing the development of segmentation and object tracking algorithms for MPEG-4 applications. We have developed a variety of algorithms as intermediate steps toward realizing the ultimate objective of semiautomatic video segmentation. Some of our developments include:

QuickSeg	Fast user-assisted segmentation
ATracker	Tracking using a deformable template model
VSnakes	Tracking for deformable objects
RAVEN	Extensible framework for developing and testing algorithms

Information about MPEG-4 is available at the ISO/IEC MPEG home page.

Projects

RAVEN: Research Architecture for Video Editing and eNcoding

Wed, 2006-11-22 13:35 — sluo

Introduction

RAVEN is a user-friendly environment for segmenting video objects. Users can segment manually, use a variety of automatic or semi-automatic plug-in algorithms, or use a combination of manual and computer-assisted methods. An integrated MPEG-4 encoder allows users to efficiently compress the video in an industry standard format.

Features

Plug-in interface allows users to easily add new segmentation and tracking algorithms to the environment
Algorithm developers can concentrate on writing algorithms instead of creating a user interface
Consistent and friendly user interface for different segmentation and tracking methods
Multiple undo/redo for single-frame editing operations
Multiple display modes for visualizing segmentations to assess their quality
Timeline display for identifying segmented video objects and the frames that have been segmented
Interactive segmentation plug-ins enable faster initialization of tracking algorithms and fast

Projects

VSnakes

Wed, 2006-11-22 13:25 — sluo

Motivation

Segment and track deformable video objects

Algorithm

Global relaxation with Viterbi algorithm
Template matching with CAPS (Correlation Auto-Predictive Search)
Scene change detection with M-ART2

View video (3.69 MB MPEG-1)

Frame 1
Frame 27
Frame 115

Projects

ATracker

Wed, 2006-11-22 12:46 — sluo

Motivation

Utilize shape information from previous frames
Enable global and local deformations during object tracking

Algorithm

Deformable template followed by local segmentation

Frame 1
Frame 15
Frame 39

Projects

MAP UWICL

Wed, 2006-11-22 11:36 — sluo

Table of Contents

Introduction

History, Goals and Features of MAP UWICL

Architecture of MAP UWICL

UWICL Functions

Additional Library Functions To Be Developed

Quality Assurance

Documentation

Support

Introduction

University of Washington Image Computing Library (UWICL) consists of a set of image computing functions for the Media Accelerated Processor (MAP). The document describes the history, goals and features of the MAP UWICL, its architecture, the image computing functions supported, the quality assurance procedure, the documentation and technical support provided.