Pre-trained object detection model (SSD) is used to summarize lengthy CCTV footage by identifying the object(s) of interest in each frame. Only the frames with the objects are kept.
Feeling supportive is the key to older people. My project help geriatric patients to cure their loneliness in today's busy world.
Learning Chinese language can be a challenging task. How about we use CNN to help you learn and distinguish similar Chinese consonants? learning and practicing made easier!
Crop an object of an image typing its name automagically
Regenerate any "Thing"
We train an autoencoder using the BOLD5000 dataset to get a transferable feature extractor, where we use the encoder to extract features and employ the decoder for an independent fMRI dataset
This assistant helps the people in many ways like remind you for your medicine, follow your command to contact anyone in your contact list.It also provide the facillity to learn everything from google
Endless AI-Generated Tees
Care4u identifies through the camera how healthy is the food you are about to eat regarding your current health condition, keeps track of nutrients, fat, cholesterol and others of what you want to eat
I am contributing to development of framework which will work on document images and convert it into understandable format for dyslexics.
Generate music with AI
Using computer vision for object detection and recognition find and collect tennis balls
French-English translations using PyTorch’s Transformer model and paper implementation of "Attention is All You Need”
Use deep learning to detect skin anomalies by taking a photo of your skin with a smartphone in order to prevent future skin diseases.
An AI Personal Training App
Automated traffic violation detector
Detecting ingredients in Thai menus
Understand Human, Predict Future
Building a deep learning model to detect and classify the birds around the world
SynaNN, inspired by synapses research in neuroscience, shows special features and high performance to apply synapses and deep learning to implement image classification with MNIST dataset by Pytroch.
Implementation of the state of the art algorithms like QuatE, ComplEx, and HOLE for learning knowledge graph embedding for relations and entities.
PyGlow is a Python package which attempts to implement Keras like API struture on PyTorch backend. It provides functionalities which supports information theoretic methods in deep learning.
Autoencoder trained on the BOLD5000 dataset
A fast and efficient framework for training hierarchical RL models in pytorch.