Kumar Selvakumaran's Projects
Homework 2 - Weakly Supervised Object Localization
Inspired by the the legendary "The Annotated Transformer" This project aims to build an intuitive implementation of the YOLO model from scratch.
Android apps that host QR marker based AR experiences. Textbooks have QR Codes that host AR experiences to better understand conscepts.
Extendable and real-time Content Based Image Retrieval
A Computer vision tutorial for aspirants who are just starting out.
Interview questions to ponder related to computer vision.
Exploring the different concepts of modern c++.
A deep dive into a emeddings.
This project implements a mechanism to use EigenCAM for object detection, in a plug-and-play, task-agnostic manner.
Hands-On Machine Learning with C++, published by Packt
the beginning to many projects
This is an archive of materials produced for an introductory class on CUDA programming at Stanford University in 2010
Home page
Learning Ros
Implementing Multi-Block Local Binary Pattern and Kalman Filter based tracking from scratch in python and C++
Learning C++ for ML
journey to fluency in PyTorch | current topics (FPNs, Multi-scale inference)
This project introduces image processing fundamentals and applications along with their implementations in C++ using OpenCV.
learning slam from the slambook2
Benchmarking Different transformers and demonstrating the pros of focal loss on imbalanced classification data
The course text for MIT 6.832 (and 6.832x on edX)
this is simple web application that is able to taking and stylize a video of the user's choice. this style will correspond to the style image that is also prompted from the user
Data-driven simulation for training and evaluating full-scale autonomous vehicles.
YOLOv5 in PyTorch > ONNX > CoreML > TFLite