Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
Paper Link : Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
Posts by Category
Paper
Image-to-Image Translation with Conditional Adversarial Networks
Paper Link : Image-to-Image Translation with Conditional Adversarial Networks
Unsupervised Representation Learning With Deep Convolutional Generative Adversarial Networks
Paper Link : Unsupervised Representation Learning With Deep Convolutional Generative Adversarial Networks
Learning from Simulated and Unsupervised Images through Adversarial Training
Paper Link : Learning from Simulated and Unsupervised Images through Adversarial Training
Conditional Generative Adversarial Nets
Paper Link : Conditional Generative Adversarial Nets
Generative Adversarial Nets
Paper Link : Generative Adversarial Nets
Do Vision Transformers See Like Convolutional Neural Networks
Paper Link : CNN vs VIT 최근 Transformer 계열의 Vision 모델들이 쏟아져 나오면서 기존의 CNN 모델들과는 어떤 차이점이 있는지 궁금했는데, 드디어!! 이 둘을 비교하는 논문이 나왔다. 본 논문에서는 다음의 관점들에서 VIT와 대표적인 CNN이라 ...
Polarized Self-Attention: Towards High-quality Pixel-wise Regression
Regression task를 해결하기 위한 대부분의 DCNN 모델은 encoder-decoder 구조로 이루어져 있다. 이러한 구조는 대부분 endcoder 구조에서 순차적으로 spatial resolution을 줄이는 동시에 channel resolution을 늘려나가고, d...
Attention is all you need
Abstract