Addressing adversarial attacks for computer vision algorithms by detecting and recovering the original data.
To make remote meetings more inclusive toward people with hearing impairment.
Optimize how buildings are organized to avoid damage using data science
An intelligent preprocessing wizard
A bot that detects plant diseases from leaf images and educates farmers on modern farming practices
FairTorch provides tools to mitigate inequities in deep learning. A unique feature of this tool is that you can add a fairness constraint to your model by simply adding just a few lines of code.
Drowsines, blind spot, emotions and attention monitor for driving or handling heavy machinery. Also detects objects at the blind spot via Computer Vision powered by Pytorch and the Jetson Nano.
CoviDash is an informative dashboard for information on COVID cases in Ontario.Using a heatmap and multiple graphs, data can be found for COVID cases in major location and date across the province.
Convert your math homework into LaTeX with just a picture!
Worlds first API for Deep Virtual Try on cloths exclusively for pandemic recovery in apparel industry. Powered by powerful PyTorch deep learning model with detailed cloth warping
All you need is an image of your receipt and LeggoDutch will help you break it down to items that can be charged to different people!
Vingo, that gives u emoji-puzzle, that mean some picture that contains this emojis in the room and u need to find that, after u need to put ur camera to the picture and that will recognize.
Move your hands to feed the Butterfly in the Virtual Butterfly Park
A Python library for fairness in PyTorch models.
Contact Tracing AI lets organizations leverage their existing technology infrastructure (e.g. security cameras) to prevent COVID-19 outbreaks efficiently, accurately, and at scale with computer vision
A interface that helps people generate neural network scripts based on inputs.
All You need is `forward` and `**` operator for functional neural network training!
Infinitybatch is an open source solution for PyTorch that helps deep learning developers to train with bigger batch size than it could be loaded into GPU RAM through a normal PyTorch train loop.
Auxiliary second-order optimization tools compatible with PyTorch.
Compress the parameters of any PyTorch model by a factor of 10 without reducing accuracy. Just call `spectral(your_model_here)` and fine-tune your model with an additional compression loss term.
It functions as an image segmentation website
More than a voice assistant, this project is the foundation for physical household robot trained to autonomously complete basic tasks
RecycleMate is a platform that makes it possible to recycle bottles in a multi-beneficial way for businesses by providing data using machine learning and rewarding individuals for recycling.
A python based package for monitoring and visualising the layers of a neural network.