伪装目标检测论文 - Part1

date
Oct 15, 2024
Last edited time
Oct 15, 2024 05:46 AM
status
Published
slug
伪装目标检测论文 - Part1
tags
DL
CV
summary
type
Post
origin
Field
Plat
(CVPR2023) Explicit Visual Prompting for Low-Level Structure Segmentations (WACV2024) CamoFocus: Enhancing Camouflage Object Detection with Split-Feature Focal Modulation and Context Refinement

(CVPR2023) Explicit Visual Prompting for Low-Level Structure Segmentations

Explicit Visual Prompting for Low-Level Structure Segmentations
We consider the generic problem of detecting low-level structures in images, which includes segmenting the manipulated parts, identifying out-of-focus pixels, separating shadow regions, and...
Explicit Visual Prompting for Low-Level Structure Segmentations
https://arxiv.org/abs/2303.10883
Explicit Visual Prompting for Low-Level Structure Segmentations
GitHub - NiFangBaAGe/Explicit-Visual-Prompt: [CVPR 2023] Explicit Visual Prompting for Low-Level Structure Segmentations
[CVPR 2023] Explicit Visual Prompting for Low-Level Structure Segmentations - GitHub - NiFangBaAGe/Explicit-Visual-Prompt: [CVPR 2023] Explicit Visual Prompting for Low-Level Structure Segmentations
GitHub - NiFangBaAGe/Explicit-Visual-Prompt: [CVPR 2023] Explicit Visual Prompting for Low-Level Structure Segmentations
https://github.com/NiFangBaAGe/Explicit-Visual-Prompt
GitHub - NiFangBaAGe/Explicit-Visual-Prompt: [CVPR 2023] Explicit Visual Prompting for Low-Level Structure Segmentations
通过统一的方法来处理在图像中检测低级结构的问题可以取得良好的效果。这些问题包括被操纵部分的分割、失焦像素的识别、阴影区域的分离以及隐藏对象的检测。虽然每个问题通常都有特定领域的解决方案，但采用统一的方法可以简化处理过程并提高准确性。
notion image
在我们的任务中，我们考虑了两种特征。一种是来自冻结的补丁嵌入特征。另一种是输入图像的高频分量。我们使用在大规模数据集上进行预训练的模型，并冻结其参数。然后，为了适应每个任务，我们调整嵌入特征，并学习每个单独图像的额外嵌入的高频分量。
notion image
notion image
notion image

(WACV2024) CamoFocus: Enhancing Camouflage Object Detection with Split-Feature Focal Modulation and Context Refinement

WACV 2024 Open Access Repository
WACV 2024 Open Access Repository
https://openaccess.thecvf.com/content/WACV2024/html/Khan_CamoFocus_Enhancing_Camouflage_Object_Detection_With_Split-Feature_Focal_Modulation_and_WACV_2024_paper.html
notion image
  1. 我们提出了特征分裂和调制（FSM）模块，该模块将主干特征分裂，以便更好地理解对象背景关系。
      2. notion image
  1. 内容细化模块 (CRM)：该模块促进特征的跨尺度语义理解。CRM 接收两个不同尺度的输入，使用双线性上采样进行逐通道拼接。在此之后，拼接的特征图遍历一系列卷积，然后是一个全局跳跃连接。
      2. notion image
      notion image
 

© Lazurite 2021 - 2025