Implementation of Neural Style Transfer algorithm by Gatys et. al. (2015).
Technologies Used: Python, Pytorch
A python program which segments an image using an unsupervised learning algorithm, K-means clustering in the RGB and HSV color space. Also, contains a naive K means implementation in Python without using any libraries.
Technologies Used: Python, OpenCV, scikit-learn, Numpy
Program to match the histograms of the two images (RGB and Hyperspectral) and perform feature based image alignment using SIFT.
Technologies Used: Python, OpenCV, scikit-image
Program to assign labels to each explants in a RGB Image.
Steps performed:
Technologies Used: Python, scikit-image, Pandas, Numpy
Python program which provides utilities to augment dataset for Computer Vision Projects.
Technologies Used: Python, OpenCV, Numpy, Pytorch
Trained the Deeplab v3+ model to segment the plant traits.
The segmented images are aligned with the hyperspectral images by calculating the homography matrix using optical flow.
The pipeline is used for the detection of transgenic tissues in plant tissue cultures.
Technologies Used: Python, Tensorflow, Scipy, Numpy
Trained the PSPNet model to segment the traits for plants growth analysis.
Achieved a 79.40% Mean Intersection over Union(IoU) on the test dataset.
Technologies Used: Python, Pytorch, Numpy, Pillow
Ongoing.
Technologies Used: Python, Tensorflow