Aug 2015 – Mar 2017 · 1 yr 7 mos · Bengaluru Area, India

May 2014 – Jul 2014 · 2 mos · Bangalore

• Aerospike is a flash-optimized in-memory open source NoSQL database.
• I worked on designing and developing a Nginx web server module for Aerospike. Purpose fo this module was to allow users to carry out database operations using REST APIs. REST APIs were written to support most common database operations which include connect() (connect to a database), get() (retrieve query), put() (insert query) and delete() (delete an item). Stress test was also done on the module using CurlLoader. My work mostly revolved around working with Aerospike C-client APIs, Nginx modules and curl loader.
• http://en.wikipedia.org/wiki/Aerospike_database
• Aerospike C client- https://github.com/aerospike/aerospike-client-c
• Repo : https://github.com/imvikash00/Nginx

May 2013 – Jul 2013 · 2 mos · Hyderabad

• TOPIC: Persuading visual attention through low-level image features
• Attention is one of the most important component of vision. It is the mechanism to rapidly focus on the selected portions of the visual input. Visual Saliency is the distinct subjective quality which makes some item stand different from other objects in its vicinity. Our attention is mostly attracted
• towards most salient object in the field of view. Now-a-days almost every device is capable of capturing images and focus on the part which we want them to get highlighted. But that thing may not stand out to be distinct all the times. Most of the time other factors in the image don’t let this thing happen. In this project, an attempt has been made to find a method based on Intensity so
• that the user defined region in the gray-scale image draws immediate attention of the viewer which the original image may fail to do so. Saliency map of an image is computed using the method of finding the degree centrality in the graph is used. Image is broken down into set of nodes and edge weight is computed based on feature difference, Cartesian distance and modulated by positional proximity. Feature value is taken to be the key point in determining the saliency of an image. Matlab implementation was done.