This repository contains the code for Comparing Deep Learning and Classical Computer Vision for Semantic Segmentation: A comprehensive analysis of cutting-edge techniques and algorithms for precise object segmentation in computer vision tasks. This work was done under the Computer Vision course at IIT Jodhpur.

Train a classification model to identify the product category, utilizing either classical computer vision or deep learning methods. Utilize a one/few-shot learning model to confirm the existence of a product and accurately classify its type.

Image processing with Matlab libraries

Detect and calculate orientation of fibers from SEM images

uderstanding raw concept of computer vision and how the maths used in computer vision

This parking application was developed during my first year master degree. The objectif of this application is track every car that enter to a specific car park.

Codes regarding the paper: Handwritten Image Detection using DCGAN with SIFT and ORB Optical Features

Interactive code for image similarity using SIFT algorithm

Lowe-style object instance recognition, using SIFT. The scale-invariant feature transform (SIFT) is a computer vision algorithm to detect, describe, and match local features in images

Computer Vision Course at the University of Utah

Advance Patch Matcher Implementation. Matching patches with high accuracy and short time conditions using simplified SIFT algorithm and RANSAC outlier filtering.

Sift based face recognition

3D scene reconstruction and camera pose estimation given images from different views (Structure from Motion)

Bag-of-words model is created and classification of images using K-Nearest Neighbour (KNN) and Support Vector Machine (SVM) is performed.

Python application for autostitching panoramic images.

stereo vision: estimate 3D vision depending on information extracted from 2D-images. 1)Feature extract, using SIFT algorithm. 2)Matching, using KNN algorithm. 3)Compute "Fundamental Matrix", using RANSAC algorithm. 4)Reconstruction. 5)Triangulation. 6)Pose disambiguation. 7)Rectification. 8)Disparity Computing.

Panorama composition with multible images using SIFT Features and a custom implementaion of RANSAC algorithm (Random Sample Consensus).

Coin identification and recognition systems may drammatically enhance the extended operation of vending machines, pay phone systems and coin counting machines. The primary purpose of this project is to develop a detector capable of finding and classifying Euro coins in images purely relying on Computer Vision based frameworks.

Content-Based Image Retrieval System using multiple images deciphers for feature extraction

Demonstration of sift algorithm to track objects and observing the effect of each parameter on performance.