Python package provinding tools for artistic interactive applications using AI

Last update: Sep 30, 2021

Overview

redrawing

Python package provinding tools for artistic interactive applications using AI

Created by ReDrawing Campinas team for the OpenCV AI 2021 Competition.

Installation

python -m pip install --upgrade pip
python -m pip install redrawing

Stages

Stages are the basic processing units of the redrawing package. They receive data objects from input channels, processes them and output data in the output channels.

Examples of stages:

Stage name	Use
OAK_Stage	Handles a OAK camera. Can be changed using OAK_Substages, like a Body detector or a Hand gesture detector
PCR_Viewer	Visualizer for image, depth and body pose data
CameraReceiver/IMUReceiver	Receives data from a smarthphone
UKF_IMU	UKF filter for orientation estimation using IMU data

Data and communication

All inter stage data communication occours using Data classes.

Examples of data classes:

Data class	Use
BodyPose	Stores body pose data (keypoints)
Image	Stores image data
Depth_Map	Stores depth data
IMU	Stores IMU data

For exchange data with client applications, the UDP_Stage can be used, using UDP protocol with JSON converted messages. We also have client libraries for receiving and decoding data:

redrawing_java: client library for Java
ReDrawing for Processing: client library for Processing, encapsulates the redrawing_java library

Authors

Legal

Submodules in ai_models/third_models submodules may be licensed under other licenses. A copy of the respective license can be found in the submodule folder.

Made with the support of "Conselho Nacional de Desenvolvimento Científico e Tecnológico" (CNPq) of Brazil

Python package provinding tools for artistic interactive applications using AI

Related tags

Overview

redrawing

Installation

Stages

Data and communication

Authors

Legal

Owner

ReDrawing Campinas

Framework for Spectral Clustering on the Sparse Coefficients of Learned Dictionaries

Social Network Ads Prediction

SeqFormer: a Frustratingly Simple Model for Video Instance Segmentation

Synthesize photos from PhotoDNA using machine learning 🌱

This project is based on RIFE and aims to make RIFE more practical for users by adding various features and design new models

Playable Video Generation

Hybrid Neural Fusion for Full-frame Video Stabilization

CN24 is a complete semantic segmentation framework using fully convolutional networks

(CVPR2021) DANNet: A One-Stage Domain Adaptation Network for Unsupervised Nighttime Semantic Segmentation

Neon: an add-on for Lightbulb making it easier to handle component interactions

Code release to accompany paper "Geometry-Aware Gradient Algorithms for Neural Architecture Search."

Turning pixels into virtual points for multimodal 3D object detection.

MicRank is a Learning to Rank neural channel selection framework where a DNN is trained to rank microphone channels.

KwaiRec: A Fully-observed Dataset for Recommender Systems (Density: Almost 100%)

SciKit-Learn Laboratory (SKLL) makes it easy to run machine learning experiments.

A library for uncertainty quantification based on PyTorch

Lighthouse: Predicting Lighting Volumes for Spatially-Coherent Illumination

A modified version of DeepMind's Alphafold2 to divide CPU part (MSA and template searching) and GPU part (prediction model)

Static-test - A playground to play with ideas related to testing the comparability of the code

Modelisation on galaxy evolution using PEGASE-HR