Overview

Lipi Toolkit (LipiTk) is a generic toolkit for online Handwriting Recognition (HWR), and packages HWR tools and algorithms with standard ink representations and HWR application interfaces.  The word lipi is Sanskrit for script.

Links

• Objectives
• Target Users
• Design Goals
• Usage
• Brief History
• License
• Components
• Contributions and Participation
• Getting Started

Objectives

The primary objective of LipiTk is to simplify the creation and deployment of HWR technology for new scripts. Other broad goals include supporting collaborative HWR R&D in academic and industrial settings, acting as a tool for user interface research, supporting commercial HWR development, promotion of standard ink representations and interfaces, promotion of sharing & reuse of tools, algorithms, code and handwriting datasets, and promotion of product and solution development.

The first incarnation of LipiTk is designed to support the recognition of isolated characters and shapes.
 

Target Users

The toolkit is intended to support different sets of users, with different requirements. However some knowledge of C++ is a prerequisite.

Design Goals

• Data-driven methodology for recognizer development – data collection, training, evaluation
• Generic, script-independent algorithms and tools
• Emphasis on robust and efficient implementation, coding standards, rigorous testing
• Multiple platform support: Minimally, windows and Linux
• Build and run-time configurability
• Extensibility: Plug in new tools and recognition algorithms for specific scripts, applications
• Standards: Standard Ink and dataset representation, recognition interfaces
   

Usage

Brief History

For more historical background, click here.

License

Except for some tools that are licensed under GPL, LipiTk is licensed under the MIT open source license, and hence there are no restrictions on commercial use, or a requirement to put modifications back. Please see license terms for details.

Components

LipiTk is intended as an umbrella for distribution of:

• the Core Toolkit, composed of common classes for representing ink, and algorithms for preprocessing, feature extraction and recognition. A number of scripts and utilities are also included.
• Standalone tools for data collection and annotation
• Demos and applications

Please see the Downloads page for details of what is presently available on each of the above fronts, and the Roadmap for what is in the pipeline.

Contributions and Participation

We hope that over a period of time LipiTk will grow to include both core toolkit components (such as new features and recognizers) as well as standalone tools and applications contributed by others.  See the Community page for more details.

Getting Started

A good place to get started is with the first few chapters of the Core Toolkit user manual, available from the Downloads page. The use cases at the end of the user manual describe the essential steps in creating a new character or shape recognizer.

Organizations - Legend