Segmentation and Detection of Work Zone Scenes
This project is concerned with the automatic detection and analysis of work zones in naturalistic roadway images.
SUMMARY OF QUALIFICATIONS
- Currently working as a full time machine learning research assistant at Virginia Tech Transportation Institute
- 3 years of experience in developing and deploying Software & Machine Learning code on cloud and distributed systems
- Experienced in Python, C++, SQL, Docker & AI/ML frameworks such as Scikit-Learn, SciPy, Apache Spark, PyTorch, Keras & TensorFlow.
PROJECTS
Super Sampling
Super Resolution using Generative Adversarial Networks GAN and Autoencoders
Speech Emotion Classification
Classifying Human Emotions using LSTMs and Transformers Using Two Feature Extraction Methodologies
Fog Computing
Analysis of The Implementation Of Internet-of-Things. Demonstrate Superiority Of Fog-Computing Compared To Cloud.
Face Anti-Spoofing
CNN Based Face Anti-Spoofing Model That Can Detect A Spoofed Image With High Accuracy.
Face Landmark Detection And Adding Filters
In This Project We Use Pythons 'dlib' Library To Find 68 Key-Points On A Face And Apply Filters To The Detected Face. One can extend this to a live face as well.
Distracted Driver Detection Using Deep Learning
Using Deep Learning Techniques To Find If A Driver Is Distracted Or Not, Using Images Of Various Drivers Captured Through A Camera Mounted In A Car.
IoT Based Student Management System
A Duplex Communication System That Is Capable Of Transmitting Information Regarding Changes In The Timetable Or Upcoming Seminar/Notices To The Lecture Theater.
IoT Based Home Automation System
A Simple Project That Incorporates IoT Using BOLT Microcontroller To Enable Home Automation. In the code I have Provided With A Way To Remotely Turn On Appliances.
PUBLICATIONS
- Characterization, Detection, and Segmentation of Work Zone Scene from Naturalistic Driving Data, 1st author Transportation Research Board (2022)
- Traffic Behavior Monitoring Around Autonomous Shuttle Using Camera Based Perception System, 1st author TRR 2021- Under Review
- Performance analysis of IM/DD vs. Heterodyne detection techniques of an earth-satellite FSO link for next generation wireless communication-2017 IEEE 13th Malaysia International Conference on Communications
- Selection Combining in Hybrid RF/FSO Systems for IM/DD and Heterodyne Detection in Varying Weather Conditions-2018 2nd International Conference on Micro-Electronics and Telecommunication Engineering
- Performance analysis of an earth-satellite intensity modulated FSO link in the presence of turbulence and varying path loss-2017 7th International Conference on Communication Systems and Network Technologies
CERTIFICATIONS
Splunk Fundamentals 7.x
Image Super-resolution
Convolution Neural Networks
Hyperparameter Tuning, Regularization and Optimization
Neural Networks and Deep Learning
Structuring Machine Learning Projects
Using Database with Python
Using Python to Access Web Data
Python Programming
Python Data Structures
Home Automation Using IoT
Robotics with Embedded C