Stacked Hourglass Networks for Human Pose EstimationThis work introduces a novel convolutional network architecture for the task of human pose estimation. Features are processed across all scales and consolidated to best capture the various spatial relationships associated with the body. We show how repeatearxiv.org GitHub - princeton-vl/pytorch_stacked_hourglass: Pytorch implementation of the E..

https://arxiv.org/abs/2103.14030 Swin Transformer: Hierarchical Vision Transformer using Shifted WindowsThis paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such asarxiv.org이번 포스팅은 2021 ICCV에 accept된 Sw..

AbstractQ-Learning algorithm의 경우 특정 조건에서 action value를 과대평가하는 것으로 알려져 있다.https://arxiv.org/abs/1312.5602 Playing Atari with Deep Reinforcement LearningWe present the first deep learning model to successfully learn control policies directly from high-dimensional sensory input using reinforcement learning. The model is a convolutional neural network, trained with a variant of Q-learning, whose inp..

Atari가 무엇일까? Atari는 비디오 게임 회사이다.그럼 어떻게 강화학습 [Reinforcement Learning]에 Atari가 도입되게 되었을 까?2013년 구글 Deepmind에서 발표한 Playing Atari with Deep Reinforcement Learning 이라는 논문이 등장하면서 시작되었다. 이 논문을 통해 보이고자 한 것은 Breakout (a.k.a 벽돌깨기) 게임을 강화학습을 통해 학습을 시켜서 인공지능이 스스로 벽돌을 깨부수는 것을 확인하고자 함이였다. 특히나 이 모델의 경우 Deep 이라는 단어에서 알 수 있듯이 딥러닝을 강화학습에 적용시킨 모델이다. DQN이라고 불리는데, 강화학습 중 하나인 Q-Learning에 Deep Learning을 접목시켜서 더 효율적인 학..

논문 출처https://arxiv.org/abs/2111.06377 Masked Autoencoders Are Scalable Vision LearnersThis paper shows that masked autoencoders (MAE) are scalable self-supervised learners for computer vision. Our MAE approach is simple: we mask random patches of the input image and reconstruct the missing pixels. It is based on two core designs. First, wearxiv.orgAbstract이 논문에서는 MAE [Masked Autoencoder]가 comput..