Computer Vision Engineer at Orbo.ai

About me

Hello reader! Thanks for coming across my profile. I am curently working as CV engineer at a software firm. I have deep expertise in statistics, Image processing, Machine Learning and Deep Learning. Apart from these i have deep interest in ethical hacking and computer architecture.

Areas of Interest

Statistics
Machine Learning
Deep Learning
Computer Vision
Natural language Processing
Data Visualization & story telling

Work Experiences

Computer vision Engineer at Orbo.ai (June 2022 - present)
Interned at Episource LLC as data science intern.
Worked at Learning evolution pvt ltd (a subsidiary of Datoin) as a full stack data scientist.

Projects

Spatial mapping of soil nutrients: It was a part of my semester long project. I along with other two members are working together to come up with a deep neural network based mapping technique that can generate a high resolution map of a given area based on their geocordinates. We have used many attention techniques that can map the spatial relationship among the neigbouring points. You can find the more details about the same here.
Automatic Number plate detection: it was a part of my deep learning course. I came up with an idea to develop a number plate recogniser for Indian Vehicle number plates. There is no any standard Indian Vehicle Number Plate dataset available online. Also Vehicle numbering system does not follow any particular convention. So, It was challenging to collect the Indian Number plate and manually annotate them. YOLOV3 model was used as a number plate detector and the detected cropped image is passed through PyTesseract for digit/number recognition. If you are curious to know more about this project, head over to the project repository.
FooDE: It was a part of my Enterprise Application Development course. I along with other three members worked together to develop a Food deliver smart web app. This project was developed in an agile environment. Trough this project, we got to learn about basic software engineering practice and developing scalable application. This project can be found here.
Multi-Document Text Summarizer: This is my personal project that I worked on during my Information retriveal course. I found a genuine interest in IR system and Knowledge mining. It was a basic project made using NLTK and ML models. More details about the project can be found here.
Smart Driver Monitoring System: A deep learning, CV and mediapipe based smart driving monitoring framework that supports head pose estimation, drowsiness detection, object detection like cell phone, wine glass etc. This project can be found here.

Mini Projects

Human Action Recognition API - Detailed instructions about this API can be found here.
IIITS-OJ - It’s a basic online Judge platfrom for online coding contest. I did It along with other three members while I was exploring web development.link
Battle of Neighbourhood - It was a part of courser capstone project.link
Tic-Tac-Toe Game - I bet, you can not defeat my AI agent🤖. Challenge my AI bot here
Nine men’s Morris - Get a vibe of old days. It’s very interesting. try here

Relevant Courses

Database Management System (Fall 2019)
Machine Learning (Fall 2020)
Information Retrieval (Fall 2020)
Data Analytics (specialization elective, spring 2021)
Deep Learning (spring 2021 )
Computer Vision ( Spring 2021 )
Big Data Analytics (specialization elective, Summer 2021)
Natural Language Processing (Fall 2021)
Data Mining (Fall 2021)
Brain Computer Interaction (Spring 2022)

Certification

Roles and Responsibilities

Currently I am the club head/president of our college’s Data Science Club(Matrix). To get more details about clubs in our college please check out the official website. For any collaboration/contests get back to us on our official club mail-id.

Club mail ID: datasience.club@iiits.in

Open-Source contribution

PyMC3 - PR1 , PR2 , PR3

Publications

I have Co-authored the article “ OCFormer: One class Transformer Network for Image classification”. Currently it is under review in IEEE-SPL and the preprint version can be found here.