Computer Vision108 [2025-1] 전연주 - RePaint: Free-Form Image Inpainting with DDPM 논문 링크: 2201.09865저자: Andreas Lugmayr, Martin Danelljan, Andres Romero, Fisher Yu, Radu Timofte, Luc Van Gool소속: Computer Vision Lab, ETH Z¨ urich, Switzerland발행일: 2022 8월 31일1. AbstractFree-form Inpainting은 이미지에서 임의의 마스크 영역을 채우는 작업으로, 기존 방법들은 특정 마스크 분포에서 훈련되어 새로운 마스크에 대한 일반화가 어렵다. 또한, 기존의 픽셀 단위 및 perceptual loss 기반 방법은 의미론적으로 자연스러운 생성을 보장하지 못한다.본 논문에서는 RePaint를 제안한다.Denoising Diffusion Probabili.. 2025. 2. 14. [2025-1] 황영희 - U-Net: Convolutional Networks for Biomedical Image Segmentation https://arxiv.org/abs/1505.04597 U-Net: Convolutional Networks for Biomedical Image SegmentationThere is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotatedarxiv.org1. U-Net 이란?이미지 세그멘테이션(Image Segmenta.. 2025. 2. 13. [2025-1] 황징아이 - Temporal Feature Alignment and Mutual Information Maximization for Video-Based Human Pose Estimation 논문 : https://arxiv.org/abs/2203.15227코드 : https://github.com/Pose-Group/FAMI-Pose GitHub - Pose-Group/FAMI-Pose: This is an official implementation of our CVPR 2022 ORAL paper "Temporal Feature Alignment and MuThis is an official implementation of our CVPR 2022 ORAL paper "Temporal Feature Alignment and Mutual Information Maximization for Video-Based Human Pose Estimation" . - Pose-Group/FAMI-Po.. 2025. 2. 8. [2025-1] 유경석 - MAISI: Medical AI for Synthetic Imaging https://arxiv.org/pdf/2409.11169v2 https://build.nvidia.com/nvidia/maisi maisi Model by NVIDIA | NVIDIA NIMMAISI is a pre-trained volumetric (3D) CT Latent Diffusion Generative Model.build.nvidia.com AbstractMAISI (Medical AI for Synthetic Imaging) : 3D 컴퓨터 단층촬영 (CT) 이미지 생성 모델Volume Compression Network : 고해상도 CT 이미지 생성Latent diffusion model : flexible volume dimensions과 voxel spacing 제공ControlNe.. 2025. 2. 8. 이전 1 ··· 4 5 6 7 8 9 10 ··· 27 다음
