본문 바로가기

Review

흥미로운 SOTA 인공지능 오픈소스들..

SOTA 란?

state of the art 의 줄임말로 SOTA 알고리즘이라고 함은 인공지능 분야에서 가장 최고 성능을 내는 알고리즘을 의미하는 단어이다.

아래 링크에서 도메인별 SOTA 알고리즘을 쉽게 찾을 수 있다

https://paperswithcode.com/sota?fbclid=IwAR16wLSr-BAcl-eZAwMVIHdsmpgBF6dN4ETR5kRGg02f_5xBBOlN6uMapo4

 

Papers with Code - Browse the State-of-the-Art in Machine Learning

7578 leaderboards • 3102 tasks • 6214 datasets • 71361 papers with code.

paperswithcode.com

 

현재 SOTA 알고리즘 오픈소스 일부

  1. TransCoder

https://github.com/facebookresearch/TransCoder

 

GitHub - facebookresearch/TransCoder: Public release of the TransCoder research project https://arxiv.org/pdf/2006.03511.pdf

Public release of the TransCoder research project https://arxiv.org/pdf/2006.03511.pdf - GitHub - facebookresearch/TransCoder: Public release of the TransCoder research project https://arxiv.org/pd...

github.com

C 코드를 Python 코드로 트랜스코딩하는 AI

하지만 현재 항상 컴파일 가능한 결과물을 내지는 않는다

 

  1. CodeGen

https://github.com/facebookresearch/CodeGen

 

GitHub - facebookresearch/CodeGen: Reference implementation of code generation projects from Facebook AI Research. General toolk

Reference implementation of code generation projects from Facebook AI Research. General toolkit to apply machine learning to code, from dataset creation to model training and evaluation. Comes with...

github.com

자연어로 프로그램의 대략적인 서술을 하면 코드를 작성해주는 AI

 

  1. ArcFace

https://github.com/deepinsight/insightface

 

GitHub - deepinsight/insightface: State-of-the-art 2D and 3D Face Analysis Project

State-of-the-art 2D and 3D Face Analysis Project. Contribute to deepinsight/insightface development by creating an account on GitHub.

github.com

사람의 얼굴을 가지고 키 포인트를 확인하며 사람을 구별하는 AI

 

  1. Pythia

https://github.com/jackroos/pythia

 

GitHub - jackroos/pythia: A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR) - GitHub - jackroos/pythia: A modular framework for vision & language multimodal research from...

github.com

그림 보고 어떤 그림인지 질문에 대답하는 (VQA) AI

 

  1. GANSpace

https://github.com/harskish/ganspace

 

GitHub - harskish/ganspace: Discovering Interpretable GAN Controls [NeurIPS 2020]

Discovering Interpretable GAN Controls [NeurIPS 2020] - GitHub - harskish/ganspace: Discovering Interpretable GAN Controls [NeurIPS 2020]

github.com

 

  1. EG3D

https://github.com/NVlabs/eg3d

 

GitHub - NVlabs/eg3d

Contribute to NVlabs/eg3d development by creating an account on GitHub.

github.com

2D 이미지를 3D로 만드는 AI

 

  1. Brings old films back to life

https://github.com/raywzy/Bringing-Old-Films-Back-to-Life

 

GitHub - raywzy/Bringing-Old-Films-Back-to-Life: Bringing Old Films Back to Life (CVPR 2022)

Bringing Old Films Back to Life (CVPR 2022). Contribute to raywzy/Bringing-Old-Films-Back-to-Life development by creating an account on GitHub.

github.com

흑백이나 오래된 영상에 색을 입히는 AI

 

  1. nvdiffrec

https://github.com/NVlabs/nvdiffrec

 

GitHub - NVlabs/nvdiffrec: Official code for the CVPR 2022 (oral) paper "Extracting Triangular 3D Models, Materials, and Lightin

Official code for the CVPR 2022 (oral) paper "Extracting Triangular 3D Models, Materials, and Lighting From Images". - GitHub - NVlabs/nvdiffrec: Official code for the CVPR 2022 (oral) pa...

github.com

사진으로 유니티 3D 객체를 만들어내는 AI (매쉬 기반)

 

  1. ADOP

https://github.com/darglein/ADOP

 

GitHub - darglein/ADOP

Contribute to darglein/ADOP development by creating an account on GitHub.

github.com

사진으로 3D 공간을 만들어내는 AI (포인트 기반)

 

  1. transfiner

https://github.com/SysCV/transfiner

 

GitHub - SysCV/transfiner: Mask Transfiner for High-Quality Instance Segmentation, CVPR 2022

Mask Transfiner for High-Quality Instance Segmentation, CVPR 2022 - GitHub - SysCV/transfiner: Mask Transfiner for High-Quality Instance Segmentation, CVPR 2022

github.com

세그멘테이션 보정 AI

 

  1. DAIN

https://github.com/baowenbo/DAIN

 

GitHub - baowenbo/DAIN: Depth-Aware Video Frame Interpolation (CVPR 2019)

Depth-Aware Video Frame Interpolation (CVPR 2019). Contribute to baowenbo/DAIN development by creating an account on GitHub.

github.com

영상을 초고속카메라 영상처럼 프레임을 늘려줌

 

  1. first-order-model

https://github.com/AliaksandrSiarohin/first-order-model

 

GitHub - AliaksandrSiarohin/first-order-model: This repository contains the source code for the paper First Order Motion Model f

This repository contains the source code for the paper First Order Motion Model for Image Animation - GitHub - AliaksandrSiarohin/first-order-model: This repository contains the source code for th...

github.com

키포인트 기반의 영상화의 단점인 개와 사람의 키포인트가 달라서 새로 학습해야함을 개선한 모델, 사진과 영상이 주어지면 사진을 영상화하여 영상에 나오는 동작을 흉내내게 됨

 

  1. mat

https://github.com/fenglinglwb/mat

 

GitHub - fenglinglwb/MAT: MAT: Mask-Aware Transformer for Large Hole Image Inpainting

MAT: Mask-Aware Transformer for Large Hole Image Inpainting - GitHub - fenglinglwb/MAT: MAT: Mask-Aware Transformer for Large Hole Image Inpainting

github.com

사진에서 특정 오브젝트 지우기, 특정 오브젝트 끼워넣기

 

  1. deepchecks

https://github.com/deepchecks/deepchecks

 

GitHub - deepchecks/deepchecks: Test Suites for Validating ML Models & Data. Deepchecks is a Python package for comprehensively

Test Suites for Validating ML Models & Data. Deepchecks is a Python package for comprehensively validating your machine learning models and data with minimal effort. - GitHub - deepchecks/deepc...

github.com

데이터셋 분포가 우리가 학습한 내용과 일치하는제 체크하는 오픈소스

 

  1. DeepFashion2

https://github.com/switchablenorms/DeepFashion2

 

GitHub - switchablenorms/DeepFashion2: DeepFashion2 Dataset https://arxiv.org/pdf/1901.07973.pdf

DeepFashion2 Dataset https://arxiv.org/pdf/1901.07973.pdf - GitHub - switchablenorms/DeepFashion2: DeepFashion2 Dataset https://arxiv.org/pdf/1901.07973.pdf

github.com

옷을 입었을 때 모습을 합성하는 AI

 

  1. Real-Time-Voice-Cloning

https://github.com/CorentinJ/Real-Time-Voice-Cloning

 

GitHub - CorentinJ/Real-Time-Voice-Cloning: Clone a voice in 5 seconds to generate arbitrary speech in real-time

Clone a voice in 5 seconds to generate arbitrary speech in real-time - GitHub - CorentinJ/Real-Time-Voice-Cloning: Clone a voice in 5 seconds to generate arbitrary speech in real-time

github.com

새로운 목소리를 별도의 학습 없이도 따라할 수 있는 AI

 

  1. AirSim

https://github.com/microsoft/AirSim

 

GitHub - microsoft/AirSim: Open source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Res

Open source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Research - GitHub - microsoft/AirSim: Open source simulator for autonomous vehicles built on Un...

github.com

자율주행차를 위한 센서 값들을 제너레이트하는 시뮬레이션 오픈소스 소프트웨어

 

 

'Review' 카테고리의 다른 글

Git flow 전략  (0) 2022.05.27
RabbitMQ 메시지 Python 에서 송수신하기  (0) 2022.05.27
패스워드 관리 툴: pass  (0) 2022.05.27