Animikh Aich's repositories
ECG-Atrial-Fibrillation-Classification-Using-CNN
This is a CNN based model which aims to automatically classify the ECG signals of a normal patient vs. a patient with AF and has been trained to achieve up to 93.33% validation accuracy.
Semantic-Segmentation-using-AutoEncoders
Lightweight and Fast Person Segmentation using Autoencoders (Trained Weights Included)
Facial-Recognition-Dashboard
A flask app to give a demo of Facial Recognition - Deployed with Live Demo.
AlexNet-Tensorflow
GPU Optimized AlexNet Implementation to train on ImageNet 2012 using Tensorflow 2.x
finger-vein-biometrics
Bio-metric Identification of a person using Finger Vein Data
OpenVINO-Real-Time-Face-Anonymizer
Real-Time Face Anonymizer Algorithm using Intel OpenVINO
linux-scripts
A collection of Linux Scripts which make it easier for me to set up and install Linux libraries
No-Code-Classification-Toolkit
Containerized Tensorflow-based image classification training utility with Streamlit-based interface designed to choose between common architectures and optimizers for quick hyperparameter tuning.
Deep-Convolutional-Background-Subtractor
End-to-end CNN-based Autoencoder that can segment any objects even if it is out of the classes present in the training set.
Facial-recognition-based-automatic-door-lock-unlock-system
FACIAL RECOGNITION BASED AUTOMATIC DOOR LOCK / UNLOCK SYSTEM
Timelapse-Creator
OpenCV based Video Timelapse Creator with Graphical User Interface
VGGNet-Tensorflow
VGGNet-Family (11, 13, 16 & 19) Implementation to train on ImageNet 2012 using Tensorflow 2.x
3D-Text2LIVE
Zero-shot, text-driven appearance manipulation on multiple views of an object to generate 3D renderings.
helmetless-rider-detection
With the exponential increase of traffic in India, the number of accidents are also increasing exponentially. Most of the reported deaths due to accidents involving two-wheeler are caused due to not wearing a helmet. There is thus a need for better implementation of safety regulation Our project helps traffic police in better law enforcement due to automation in the detection of law-breaker, that is, riders without helmets. The project is based on image processing and deep learning. The algorithm built will classify the two-wheelers [1-2] in a moving traffic video, will extract head portion of the person/s on two-wheelers, following which, it would detect if he is wearing a helmet or not [3] and if the person is not wearing the helmet it will also extract the number plate [4-6] of two-wheeler. After extracting number plate, it will process it to get the registration number and it will report to RTO office. If the report comes for more than three times, strict actions may be taken against the registered person. This project can ease the burden of traffic police and can also save many lives. It will be free from human-error like, if there is a lot of traffic the policemen standing on that spot are unable to manage and punish all of them. Sometimes they increase their speed and do not stop if the officers call them. This kind of loop-holes can be eliminated as it directly takes the number of the two-wheeler from the video and reports it to concern office.
Invisible-Cloak
A program that uses Colour Segmentation using OpenCV and makes a specific colour disappear.
Image-Super-Resolution-Autoencoders
Size Invariant Image Super Resolution using Autoencoders
NeRF-Editing-GP
A computationally Efficient Implementation of NeRF-Editing with NeRF Relighting
portfolio-website
Personal Portfolio Website
animikhaich
GitHub Profile Summary
Demystifying-CIFAR-10
Using Every Possible Algorithm to Understand the Dynamics of CIFAR-10
ds-aclu-police-budget-analysis
CS 506 Police Budget Analysis
ElevateMind-Mental-Health-LLM
Mini-LLaVA
FastSAM
Fast Segment Anything for Py37
GridBlur-Improved-GridMask
GridBlur: An improved version of GridMask Data Augmentation
keras-vggface
VGGFace implementation with Keras Framework
mmcv
OpenMMLab Computer Vision Foundation
regnet
Official PyTorch implementation of the TIP paper "Generating Visually Aligned Sound from Videos" and the corresponding Visually Aligned Sound (VAS) dataset.
samplernn-pytorch
PyTorch implementation of SampleRNN: An Unconditional End-to-End Neural Audio Generation Model
VTU-Result-Downloader
A GUI App to Download Results from VTU's Website