jrbtaylor

ActivityNet

2016 ActivityNet action recognition challenge. CNN + LSTM approach. Multi-threaded loading.

conditional-pixelcnn

A PyTorch implementation of Conditional PixelCNNs

visual-attention

Torch implementation of "Multiple Object Recognition with Visual Attention" on Kaggle Cats vs Dogs dataset

ampclone

3D-Deepbox

3D Bounding Box Estimation Using Deep Learning and Geometry (MultiBin)

Combining-EfficientNet-and-Vision-Transformers-for-Video-Deepfake-Detection

Code for Video Deepfake Detection model from "Combining EfficientNet and Vision Transformers for Video Deepfake Detection" available on Arxiv and was submitted to ICIAP 2021.

domain-prior

Demonstrates the use of a PixelCNN to detect out-of-domain examples as a pre-filtering for an application-specific CNN

faceshifter

Unofficial PyTorch Implementation for FaceShifter (https://arxiv.org/abs/1912.13457)

FaceSwap

Swap face between two photos.

GANDCTAnalysis

Code for the ICML 2020 paper: Leveraging Frequency Analysis for Deep Fake Image Recognition.

higher-order-rnn

Investigate the role of higher-order terms in recurrent nets to conceptually link LSTM gates, mi-RNNs and hyperRNNs

Instagram-API-python

Unofficial instagram API, give you access to ALL instagram features (like, follow, upload photo and video and etc)! Write on python.

kaggle-uns

Kaggle Ultrasound Nerve Segmentation Challenge

keras

Deep Learning library for Python. Convnets, recurrent neural networks, and more. Runs on Theano and TensorFlow.

ortho-rnn

Investigating a soft constraint on the recurrent weight matrix orthogonality. Comparison to gradient clipping on sequential MNIST.

pytorch_pixelcnn

A PixelCNN for PyTorch w/ experiments on MNIST

recurrent_pytorch

rnn

Recurrent Neural Network library for Torch7's nn

story-generator

Train a recurrent language model on a corpus of short stories and then generate new short stories

tensorflow

An Open Source Machine Learning Framework for Everyone

trace-rnn

Test the idea of using eligibility traces to monitor neuron activations and scale the updates accordingly