Výsledky vyhledávání - "masked image modeling"

Akademický článek

Consistency Regularization Based on Masked Image Modeling for Semisupervised Remote Sensing Semantic Segmentation

Autor: Miaoxin Cai, He Chen, Tong Zhang, Yin Zhuang, Liang Chen

Publikováno v: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 17, Pp 17442-17460 (2024)

Semisupervised semantic segmentation aims to effectively leverage both unlabeled and scare labeled images, reducing the reliance on labor-intensive pixel-level labeling for extensive training processes. The leading semisupervised learning method, con

Externí odkaz: https://doaj.org/article/98742893893c4942b8c7fc8b10a563d1

Zobrazit plný text záznamu

Akademický článek

Transferable adversarial masked self-distillation for unsupervised domain adaptation

Autor: Yuelong Xia, Li-Jun Yun, Chengfu Yang

Publikováno v: Complex & Intelligent Systems, Vol 9, Iss 6, Pp 6567-6580 (2023)

Abstract Unsupervised domain adaptation (UDA) aims to transfer knowledge from a labeled source domain to a related unlabeled target domain. Most existing works focus on minimizing the domain discrepancy to learn global domain-invariant representation

Externí odkaz: https://doaj.org/article/592d3baf5c7e463197681bf45348e27d

Zobrazit plný text záznamu

Akademický článek

Masked Image Modeling Auxiliary Pseudo-Label Propagation with a Clustering Central Rectification Strategy for Cross-Scene Classification

Autor: Xinyi Zhang, Yin Zhuang, Tong Zhang, Can Li, He Chen

Publikováno v: Remote Sensing, Vol 16, Iss 11, p 1983 (2024)

Cross-scene classification focuses on setting up an effective domain adaptation (DA) way to transfer the learnable knowledge from source to target domain, which can be reasonably achieved through the pseudo-label propagation procedure. However, it is

Externí odkaz: https://doaj.org/article/b9aca1838a3649a196410a63ff06e01c

Zobrazit plný text záznamu

Plný text ve formátu HTML

Akademický článek

Masked Autoencoders in Computer Vision: A Comprehensive Survey

Autor: Zexian Zhou, Xiaojing Liu

Publikováno v: IEEE Access, Vol 11, Pp 113560-113579 (2023)

Masked autoencoders (MAE) is a deep learning method based on Transformer. Originally used for images, it has now been extended to video, audio, and some other temporal prediction tasks. In the field of computer vision, MAE performs well in classifica

Externí odkaz: https://doaj.org/article/76f70a1ab61048a9a49f27e20e803fc4

Zobrazit plný text záznamu

Akademický článek

Object-Centric Masked Image Modeling-Based Self-Supervised Pretraining for Remote Sensing Object Detection

Autor: Tong Zhang, Yin Zhuang, He Chen, Liang Chen, Guanqun Wang, Peng Gao, Hao Dong

Publikováno v: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 16, Pp 5013-5025 (2023)

Masked image modeling (MIM) has been proved to be an optimal pretext task for self-supervised pretraining (SSP), which can facilitate the model to capture an effective task-agnostic representation at the pretraining step and then advance the fine-tun

Externí odkaz: https://doaj.org/article/415638424ffe4b1cb98e0b51f3e25558

Zobrazit plný text záznamu

Akademický článek

A Novel Method for Monocular Depth Estimation Using an Hourglass Neck Module

Autor: Seung-Jin Oh, Seung-Ho Lee

Publikováno v: Sensors, Vol 24, Iss 4, p 1312 (2024)

In this paper, we propose a novel method for monocular depth estimation using the hourglass neck module. The proposed method has the following originality. First, feature maps are extracted from Swin Transformer V2 using a masked image modeling (MIM)

Externí odkaz: https://doaj.org/article/62ca39011d6d46e0aa2a9b112b3ccb3f

Zobrazit plný text záznamu

Plný text ve formátu HTML