lolz0r

ganless-hd

The basic idea is to utilize a U-Net, modified perceptual loss (pearson instead of MAE), learned basis functions, and "mean teacher" training in order to synthesize images of high quality without the normal troubles of training a GAN.

agent-framework

An LLM based tool-using neural agent framework in the ReAct paradigm, written in javascript.

active-learning

agent-maker

ceviche

:shrimp: FDFD and FDTD with Automatic Differentiation

cmr

Project repo for Learning Category-Specific Mesh Reconstruction from Image Collections

ConvLSTM

Spatio-temporal video autoencoder with convolutional LSTMs

Detectron

FAIR's research platform for object detection research, implementing popular algorithms like Mask R-CNN and RetinaNet.

EvoGrad

ffmpeg-python

Python bindings for FFmpeg - with complex filtering support

IIC

Invariant Information Clustering for Unsupervised Image Classification and Segmentation

lolz0r

mean-teacher

A state-of-the-art semi-supervised method for image recognition

paac

Open source implementation of the PAAC algorithm presented in Efficient Parallel Methods for Deep Reinforcement Learning

poincare-embedding

Poincaré Embeddings for Learning Hierarchical Representations (https://arxiv.org/abs/1705.08039) in PyTorch

pyscatwave

Fast Scattering Transform with CuPy/PyTorch

pytorch-coviar

Compressed Video Action Recognition

pytorch-es

Evolution Strategies in PyTorch

pytorch-image-comp-rnn

PyTorch implementation of Full Resolution Image Compression with Recurrent Neural Networks

stereo-magnification

Code accompanying the SIGGRAPH 2018 paper "Stereo Magnification: Learning View Synthesis using Multiplane Images"

Ultraino

Acoustic Field simulation for phased-array controllers

w-net

w-net: a convolutional neural network architecture for the self-supervised learning of depthmap from pairs of stereo images.