Digitization of 15th Century European Books Archives in differnt languages (Latin, French, German, etc.) into searchable text (OCR). Developing a complete end-to-end document processing pipeline which includes, preprocessing, layout analysis and OCR. The main challenges are: image degradation and historical (unavailable) fonts.

The next-generation End-to-End Solution for Architects During Early Design Phases. While most of the established tools in architecture focus on creating concrete floorplans, Archistant supports the architect in developing ideas and getting inspiration.

The intelligent and anticipated physics book, which we are developing in the HyperMind project with the help of sensors like eye trackers, Infrared cameras, etc., will become a dynamically adaptive personal textbook and enable individual learning. The combination of a digital textbook with activity detection (for example workload, understanding, and interests of students) offers the possibility of dynamically generated content individually for each student to verify demand for specific content.

Pakistan's principal natural resources are cultivable land and water. Agriculture accounts for 25% of GDP and employs about 50% of the labor force. The hundred years old canal based irrigation system is inefficient and incompatible, and wastes about 50% of 105 maf water. The goal of this project is to develop a smart irrigation system that not only saves water but also increase growth.

In the STEM (Science, Technology, Engineering, and Mathematics) subjects there is a great demand for specialists. In addition, the rate of study breaks is especially high in these subjects mainly because of difficulty in understanding practical problems. In this project we have been developing intelligent and interactive setup (using eye tracking and head mounted displays) to inspire students towards STEM studies through presenting the practical learning examples as clear, simple and interactive as possible.

Introducing a generic digital mail-room web-service DOC-TOR (Document Investigator) that defines a complete automated process of document analysis for various different business use cases (like forms, invoices, emails as well as legal, technical and heterogeneous documents) and personal use cases. It is equipped with cutting edge document processing, classification and information extraction techniques.

The OCRopus project is an on-going effort to create a high-performance OCR system for both printed and handwritten text, and to develop novel and robust algorithms for document image preprocessing, page segmentation, text recognition, and statistical language modeling.

Decapod is a project focused on building a low-cost digitization solution that will allow for rare materials, materials held in collections without large budgets, and other scholarly content to be digitized into a high-quality PDF format. This project will work to incorporate the hardware and software necessary to accomplish this goal.

A substantial impediment to socio-economic development, good governance and effective decision-making for rural Pakistan is the absence of administrative, demographic and thematic maps. From official products of Survey of Pakistan to crowd-sourced Google Maps, rural Pakistan mostly is a large empty void. Detailed maps that can support decision-making are simply non-existent, thus limiting capacity across government departments for tasks such as locating a new school, coordinating response to natural disasters or simply comparing indicators across communities. The colonial-era village maps, hand-drawn through field surveys decades ago, are stored in district record rooms across the provinces, and contain detailed mapping of each mauza and its land parcels. The project presented a practical solution to change this state of affairs, through scalable and automatic vectorization and mosaicing of these old colonial-era village (mauza) maps. .