Hrishikesh Pawar

I am a masters student in Robotics Engineering Program student at Worcester Polytechnic Institute (WPI). My career interests and expertise lie in tackling real-world problems by developing end-to-end solutions enabling machines to perceive and interact with their environment intelligently. I am currently working as a Graduate Student Researcher at the Perception and Autonomous Robotics Group, focusing on high-speed autonomous drone navigation using Neuromorphic Vision Sensors .

Prior to grad school I spent 3 amazing years at Adagrad AI , a fast paced startup, delivering production-grade Computer Vision solutions for global clients. I was involved in solving challenges in real-time video analytics, such as OCR, license plate recognition, pose estimation, person fall detection, and semantic segmentation.

I am currently looking for Co-Op and Full-time opportunities in Computer Vision, Perception and ML starting Fall 2024..

Email  /  Github /  LinkedIn  /  Resume   

Relevant Coursework: Computer Vision, Deep Learning, Vision-Based Robotic Manipulation, Motion Planning

Affiliations: PeAR Lab, MER Lab

I spent my undergraduate days working in the Combustion and Driverless Formula Student Team designing, building and racing Formula-3 prototype racecars at various national and international competitions. My four-year stint in the team allowed me to implement several concepts from my coursework, building a solid foundation in Mechanical Engineering fundamentals.

As an AI Software Intern at Nobi, I led the development of smart ceiling lamps focused on real-time fall detection and emergency response for elderly care. I developed a rotation-aware detection model by using Swin Transformer backbones, enhancing the model’s performance in real-world scenarios. I also worked with vision-language models like LLaVA and CLIP, using LoRA for fine-tuning to improve task generalization. Additionally, I automated the entire deployment pipeline using Jenkins, Kubernetes, and Docker, streamlining the process from model training to real-world application.

Developed hardware accelerated Computer Vision products addressing crucial real-world problems.

Gate-Guard: The focus was on creating an edge-based boom barrier system leveraging Automatic License Plate Recognition (ALPR) for vehicle access control. I was involved in developing data collection pipelines, model training, and deployment tailored to lightweight object detection models like Yolo-X and Yolo-v5, optimized for Nvidia Jetson TX2. Beyond model implementation, I developed interactive analytics and monitoring services using Django, Azure, WebSockets, Kafka, Celery, and Redis to ensure real-time data processing and system scalability.

As part of my ongoing research at PeARlab, WPI, I am developing adaptive optics with event cameras for navigation, focusing on dynamically adjusting focal length and aperture to gather depth cues in real-time. I am also implementing a reinforcement learning (RL) policy to control optical parameters, aimed at improving navigation efficiency and obstacle avoidance. Additionally, I am building a custom simulator using Gaussian Splats to generate high frame-rate, realistic frames and events, enabling Hardware-In-The-Loop (HITL) testing for more accurate performance evaluation.

Utilising depth images and point clouds to manipulate the waste stream using a robotic arm, and uncover occluded and covered objects

Project Link

Implemented the Neural Radiance Fields (NeRF) technique for synthesizing novel views of scenes. This project leverages deep neural networks to model the volumetric scene function, encoding the density and color of points in space as a function of viewing direction and location.

Applied gradient descent optimization to adjust network parameters, achieving photorealistic image synthesis from sparse input views by accurately interpolating the light field.

Developed a the Classical Structure from Motion (SfM) pipeline to reconstruct 3D structures from sequences of 2D images. This project integrates key techniques such as feature detection, matching, motion recovery, and 3D reconstruction.

Utilized essential matrix computation, bundle adjustment, and triangulation methods to accurately estimate 3D points and camera positions, demonstrating the core principles of SfM in computer vision.

Implementated the seminal work of Zhengyou Zhang from scratch to estimate the camera intrinsics and distortion parameters.

Used SVD and MLE for estimating the camera calibration parameters.

Implemented Feature Detection followed by Adaptive Non-Maximal Suppression (ANMS) to ensure even distribution of corners across images, enhancing panorama stitching accuracy..

Utilized Feature Matching and Robust Homography estimation using RANSAC, further refining match quality. Employed image blending startegies like alpha blending and poissons blending.

Implemented a edge detector which works by searching for texture, color and intensity discontinuities across multiple scales..

Essentially it is a simpler implementaion of Pablo Arbelaez's paper.

Implemented AdaAttN for diverse style application on images with text-based image segmentation using CLIPSeg.