Official implementation of PointBeV: A Sparse Approach to BeV Predictions

This mainly focus on the perspective transform of an image using Opencv

Project Idea is to train an object detection model which can detect humans in a video and results in all the bounding boxes with humans upon which I have calculated the distance between boxes. Bounding boxes in closer than a threshold value are classified with different color. Also created Bird's eye view using which we can avoid overlapping.

This repository contains a ROS node for autonomous lane following. The system processes camera images, applies a Bird's-Eye-View (BEV) transformation, and detects lane lines using a robust Sliding Window technique. A target path is then planned, and the Pure Pursuit algorithm calculates the necessary steering angle to guide the vehicle.

This tool has four features, which are detecting humans in the frame using deep learning algorithms (faster-rcnn and yolo), calculating the distance between every human who is detected in the frame (scale factor and homography), showing how many people are at high, low and not at risk and generating notifications to warn if anyone is found viola…

This Python project detects and classifies vehicles (bike, car, truck) from a video feed using OpenCV. It applies a bird’s eye view transformation to better track and analyze vehicle motion within a defined Region of Interest (ROI).