Safety First utilizes state of the art computer vision to determine whether you are wearing a hard hat or not! I'm using fastai, Flask for the backend, and Monday's webhook integration to make this.
Software para Soporte del Proceso de Reposición de Carnes de la Rede BonArea
AI powered Blind Companion
A web app which provides a summary and also generates questions from the huge text given in the form of images, URL links of pages or directly copy-pasting the text to be summarized.
Through the integration of NLP and Deep Learning, and using a feed-forward neural network as a learning model, we created a chatbot to assist communities: Rona
Fighting against misinformation and hate speech on Twitter using scalable semantic matching and user-specific topic filtering systems based on personal triggers.
We recommend other twitter profiles you definitely want to follow.
#HowAreYouTweeting promotes self awareness and mental health conciousness for Twitter users during the pandemic and beyond.
Fine-tuned deep learning models for text classification
Our idea is to generate shopping lists for users based on past purchases and when they took place. Users input past lists, data is stored in firebase, and a machine learning algorithm analyzes it.
SMARTLook is a chrome extension that helps blind people navigate the web more easily. It uses machine learning to generate captions(alt property in html img) for images that the TTS software can read.
Have you ever walked by a shop and wondered how a certain item would look on you? Now you don't have to with MyWardrobe. MyWardrobe is a website which lets you see yourself in clothing you shop for
Modern solution to track calorie intake. Don't waste time manually adding meals to some food tracking app, just take a picture and let AI do the rest.
AI based tool that provides insights on tasks at hand to remote workers.
Developing a Convolutional Neural Network for facial recognition, using face images of active players from the 2019-2020 NBA season.
Learning texture representations: Learn high quality textures of 3D data to enable learning of probabilistic generative models for texturing unseen 3D models
Use Clinical Notes to predict icu outcomes
Merging Deep Neural Networks & Random Forests :)
In this work, we propose an improved version of Cutout, which utilizes per-pixel importance scores created by a model interpretability method, Grad-CAM. We call this method Gutout.
A python desktop application that allows you to do image inpainting by directly drawing on it.
A rapid Covid19 test that’s available to everyone with an internet connection and a device that records audio. Our machine learning algorithm will analyze your cough and determine if you have Covid.
We are reimplementing a neural network architecture that can detect the early onset of depression through users’ social media text.