PDF to Text Extraction

by Nandyala Pavan Kumar

Extracting text from PDF documents is a common pre-processing task for text analysis and NLP work. The main challenges tools face in extracting content from PDF files is that PDFs are composed of text, graphics and tabular structures encoded in a form designed for printing.

Warranties for smart contracts

by Srikumar Subramanian et al

The distributed ledger protocol used by blockchains has resulted in systems where we do not have to place trust in particular parties involved in maintaining these ledgers. Moreover the ledgers are programmable with “smart contracts” - transactors whose state changes are recorded and validated on the blockchain. A collection of smart contracts describes a system that is expected to ensure certain invariants relevant to the domain are upheld. For example, an election system is expected to maintain voter confidentiality. While the code that describes these “smart contracts” is open for anyone to read, those who’re participating in the systems run by these smart contracts are not in general competent to evaluate them, with the OSS community being the sole eyes on the contracts being deployed. In this post, I examine how these smart contracts can provide “warranties” that are easier to ratify and describe clear and automatic consequences of violating the warranted properties.

i-tagger - DL models for sequence tagging

by Mageswaran Dhandapani
i-tagger - Neural Networks based Deep Learning models and tools for sequence tagging. Developing models to solve a problem for a data set at hand, requires lot of trial and error methods. With current projects, we find a difficulty with supporting different datasets and models in a modular way. i-tagger helps with easing preprocessing, training and prediction. https://github.com/Imaginea/i-tagger

A Journey to the Virtual World

by Ashok Regar et al

Welcome to our Christmas post on our work on data visualization in immersive media!

Technology is evolving rapidly and it has become a part of our daily lives. One of the top trending technologies today is virtual reality (VR). And we, at Imaginea Labs, are probing into it with gusto.


by Manoj Kumar
As of year 2017 we have many ways to download raw bitcoin data. If we want to use this bitcoin transaction data we need to first understand, extract and represent the data in some meaningful format. To achieve this it involves use of resources and time. To make your life easier we made efforts in providing the extracted bitcoin data in CSV format which can be used directly for an individual’s requirement.

Blockchain apps must be closed systems

by Srikumar Subramanian

Blockchain tech, especially smart contracts, are the hot new “internet”. Post the creation of Bitcoin, we’ve seen the rise of the public smart contract system Ethereum and several “private” systems like The Linux Foundation’s Hyperledger. These distributed ledgers have become the hot new foundation to build apps on top of, leveraging the additional trust that they are supposed to provide by virtue of their distributed nature.

Interactive Fluid Simulation

by Ramakrishnan Mohan

Realistic physics-based simulations are a great way to engage a user in a gestural interactive system. Nowadays we witness such systems in shopping malls that allow us to try on clothes virtually, or play games. Among such simulations, it has been found that interacting with fluids such as water has a calming effect. We wanted to replicate this feeling digitally as an interactive wall. Capturing how water moves using a finite element simulation of a water surface turned out to be an interesting challenge in our exploration.

In this post, we share some interesting results from this experiment in which we implement a fluid simulation system towards such a water surface that a user can play with, as well as a few alternative, colourful visualizations of disturbed viscous fluids. The interactions were effected using a Kinect2 and the visualizations were projected on a wall for an immersive effect as shown in the following videos.

Deep Type

by Manoj Kumar et al
At Imaginea, we run a social network for typoholics called Fontli as our designers have a passion for the field. Folks share typography that they catch in the wild or work that they’ve created themselves. Members ask others for font identification and tips, and tag what they’re able to identify themselves. Given that we’re into typography, we would love to have a system where we can take a picture of some type and apply it to text of our own choice!

Typography Detection

by Uttam Erukala et al

Keeping undesirable content out of social networks and communication channels is a common problem. Our email systems today have sophisticated “spam filters” thanks to which we’re protected from much harm and waste of time. The problem of spam is particularly harsh in niche social networks and interest groups which are small and sensitive to disruption. We run one such niche social network for typography enthusiasts called Fontli and we like to protect our dear typographers from content that they’re not interested in - which is everything that isn’t typography. The problem is that this is hard … even for humans!

In this post, we talk about a filter we recently developed and deployed to reduce and flag incidences of non-typographic content on Fontli, using a deep convolutional neural network based image classifier. We’ve had modest success and faced some intriguing situations and results along the way.