Experience

IDeeplite provides an AI-Driven Optimizer to make Deep Neural Networks faster, smaller and energy-efficient from cloud to edge computing.

Intel Labs, the research arm of Intel Corporation, is inventing tomorrow's technology to make our lives easier.

• Working in the intersection of Deep Learning, Computer Vision and Computer Graphics in the Graphics Research Team at Intel

• Working on Neural Rendering for embedding learnable components in traditional graphics pipeline

• Working on different aspects of Temporally Consistent Neural Style Transfer

• Worked on developing a in-house barcode recognition network. As a part of literature survey, both classical and deep learning methods were explored. Due to lack of real data, extensive research was done in simulating data which lead to developing a barcode recognition network with about 95% accuracy.

• Ported a SOTA object detection network onto an embedded device. Did thorough optimizations using TensorRT getting 20 FPS and only 1.5 GB memory footprint.

As a part of Vision-enabled sensing group under the supervision of Prof. Thambipillai Srikanthan and Asst. Prof. Lam Siew Kei, I was involved in building computer vision systems for moving towards the ultimate public transportation system.

• The Problem was to label an unlabelled dataset with bounding box around the vehicles without any training data.

• Proposed a model for Unsupervised Domain Adaptive Object Detection for safe and efficient movement of pedestrians glued to their phones.

• Improved the results of the current state-of-the-art by incorporating a special NMS(Non Max Suppression) and post-processing on custom NTU road dataset.

Sally Robotics is an 'Autonomous Vehicles'​ research group by robotics researchers at the Centre for Robotics & Intelligent Systems(CRIS), BITS Pilani and spearheaded by Prof Bijay Kumar Rout. The research group aims to develop a truly autonomous car for the challenging Indian Roads.

I'm currently associated with the development of the computer vision subsystem to achieve complete spatial perception for the Autonomous Vehicle.

I worked in the Infinity Labs(R&D Lab) of UST Global, Trivandrum which takes in research grade projects and transforms them to scalable projects.

1)U-Store Worked to develop a next generation store using Deep Learning and Computer Vision. It will be entirely cashier-less and will begin the era of Smart-Stores(Just like smartphones).

>Worked on the implementation and testing of Facial and Object Recognition using Deep Learning libraries like Keras and TensorFlow

>Integrated the face recognition and object recognition module with the web portal using flask.

2)InfiLoan A product which will simplify the online car loan process for both the customer and the financial institutions. We used Machine Learning algorithms to predict the approximate financial status of the customer using data retrieved from their social media accounts.

This work lead to a Research Paper titled Social Network Analysis using Data Segmentation and Neural Networks, IRJET, June 2018

3)Crowd Detection at a large cafeteria using Computer Vision and predicting the number of empty chairs left during peak rush hours for saving the precious time of employees. We used multiple approaches to the problem like using openCV and YOLO (You Only Look Once)object detection algorithm.

4)Fault Prediction analysis in Aircraft tires using various machine learning models and building a predictive model based on multiple features. It will help in automating the supply chain management system to buy new equipment beforehand. This will help to avoid any delay in acquiring parts and prevent subsequent loss in revenue.

• Developed user-friendly UI for various clients using JavaScript, jQuery,HTML/CSS and BootStrap. SiteCore was used as the CMS(Content Management System).

• Worked with the business and marketing teams to understand the perspective of the users to launch easy-to-use and scalable products