显著性检测
文章平均质量分 96
有为少年
一步步,一点点
展开
-
CVPR 2023 | Texture-guided Saliency Distilling for Unsupervised Salient Object Detection
无监督显著性目标检测任务的常用策略是伪标签手段. 伪标签中会存在大量的噪声. 如何处理带噪标签是无监督显著性目标检测任务工作的一大重点. 现有方法专注利用有着更加可靠标签的容易样本, 但是忽略了难样本中有价值的知识. 这篇文章中关注与同时挖掘难易样本中的显著性知识.原创 2023-04-20 16:07:33 · 2468 阅读 · 5 评论 -
显著性目标检测之Learning to Promote Saliency Detectors
Learning to Promote Saliency Detectors论文阅读旧文重发https://github.com/lartpang/Machine-Deep-Learning缩写标注:SD: Saliency DetectionZSL: Zero-Shot Learning关键内容:没有训练直接将图像映射到标签中的 DNN。相反,将 DNN 拟合为一个嵌入函数,以将像素和显著/背景区域的属性映射到度量空间。显着/背景区域的属性被映射为度量空间中的锚点。然后,在该空间中原创 2021-10-07 14:51:08 · 1292 阅读 · 0 评论 -
CoSOD之Re-thinking Co-Salient Object Detection
Re-thinking Co-Salient Object Detection文章目录Re-thinking Co-Salient Object DetectionCoSOD是什么应用前景现有数据集存在的问题CoSOD的评估现有评估方式的局限CoSOD与SOD评估方式的差异本文的贡献CoSOD3kCoEG-NetCo-attention Projection for Co-saliency Learning实验结果讨论和建议相关链接原始文档:https://www.yuque.com/lart/pap原创 2020-12-07 19:46:51 · 1389 阅读 · 0 评论 -
CoSOD之ICNet: Intra-saliency Correlation Network for Co-Saliency Detection
ICNet: Intra-saliency Correlation Network for Co-Saliency Detection文章目录ICNet: Intra-saliency Correlation Network for Co-Saliency Detection前言整体框架提取intra Cues提取Inter CuesRearranged Self-Correlation Feature实验细节失败案例分析相关链接原始文档:https://www.yuque.com/lart/pape原创 2020-12-07 19:28:39 · 720 阅读 · 0 评论 -
显著性目标检测之Global Context-Aware Progressive Aggregation Network for Salient Object Detection
Global Context-Aware Progressive Aggregation Network for Salient Object Detection文章目录Global Context-Aware Progressive Aggregation Network for Salient Object Detection主要工作主体结构FIASRHAGCF损失函数实验细节相关链接原始文档:https://www.yuque.com/lart/papers/kyxtc1AAAI 2020原创 2020-05-28 00:30:57 · 1814 阅读 · 0 评论 -
显著性目标检测之F3Net: Fusion, Feedback and Focus for Salient Object Detection
F3Net: Fusion, Feedback and Focus for Salient Object Detection文章目录F3Net: Fusion, Feedback and Focus for Salient Object Detection说在开头主要工作主要结构损失函数实验细节相关链接原始文档:https://www.yuque.com/lart/papers/mq2x1yAAAI 2020说在开头这篇文章的代码用了一些很有用的trick,其中我觉得最重要的就是多尺度训练。原创 2020-05-28 00:24:53 · 2744 阅读 · 2 评论 -
显著性目标检测之Progressive Feature Polishing Network for Salient Object Detection
Progressive Feature Polishing Network for Salient Object Detection文章目录Progressive Feature Polishing Network for Salient Object Detection主要工作主要结构实验细节相关链接原始文档:https://www.yuque.com/lart/papers/gcctz1AAAI 2020主要工作本文还是从多层级特征利用的角度入手,提出了一个Progressive Fea原创 2020-05-28 00:19:12 · 1174 阅读 · 0 评论 -
显著性目标检测之Shifting More Attention to Video Salient Object Detection
Shifting More Attention to Video Salient Object Detection文章目录Shifting More Attention to Video Salient Object Detection主要改进数据集层面模型层面实验设置实验比较相关链接原始文章:https://www.yuque.com/lart/papers/vrkwzi好久没有更新了,...原创 2020-04-30 12:27:07 · 1373 阅读 · 0 评论 -
显著性目标检测之Towards High-Resolution Salient Object Detection
Towards High-Resolution Salient Object Detection文章目录Towards High-Resolution Salient Object Detection主要贡献针对问题主要方法GSN&LRNAPSGLFN实验细节相关链接原始文档: https://www.yuque.com/lart/papers/fgwcg5主要贡献提供了第...原创 2019-12-22 15:15:56 · 2230 阅读 · 0 评论 -
显著性目标检测之Selectivity or Invariance: Boundary-Aware Salient Object Detection
Selectivity or Invariance: Boundary-Aware Salient Object Detection文章目录Selectivity or Invariance: Boundary-Aware Salient Object Detection主要贡献针对问题主要方法实验细节参考链接原始文档: https://www.yuque.com/lart/papers/...原创 2019-12-22 15:03:43 · 1704 阅读 · 0 评论 -
显著性检测之SAC-Net: Spatial Attenuation Contextfor Salient Object Detection
SAC-Net: Spatial Attenuation Contextfor Salient Object Detection文章目录SAC-Net: Spatial Attenuation Contextfor Salient Object Detection主要工作主要结构实验细节一些讨论参考链接原始文档:https://www.yuque.com/lart/papers/sacnet...原创 2019-10-06 22:08:08 · 1148 阅读 · 9 评论 -
显著性检测之Semi-Supervised Video Salient Object Detection Using Pseudo-Labels
Semi-Supervised Video Salient Object Detection Using Pseudo-Labels文章目录Semi-Supervised Video Salient Object Detection Using Pseudo-Labels主要工作主要结构Flow-Guided Pseudo-Label Generation Model(FGPLG)Video s...原创 2019-08-22 22:09:35 · 2413 阅读 · 0 评论 -
显著性检测之Contour Loss: Boundary-Aware Learning for Salient Object Segmentation
Contour Loss: Boundary-Aware Learning for Salient Object Segmentation文章目录Contour Loss: Boundary-Aware Learning for Salient Object Segmentation网络结构Contour LossHierarchical Global Attention Module实验细节参...原创 2019-08-22 22:03:41 · 3233 阅读 · 9 评论 -
显著性检测之A Bi-directional Message Passing Model for Salient Object Detection
A Bi-directional Message Passing Model for Salient Object Detection文章目录A Bi-directional Message Passing Model for Salient Object Detection主要工作网络结构实验细节相关链接原始文档:https://www.yuque.com/lart/papers/bmp...原创 2019-08-22 21:47:14 · 1514 阅读 · 0 评论 -
显著性检测之Cascaded Partial Decoder for Fast and Accurate Salient Object Detection
Cascaded Partial Decoder for Fast and Accurate Salient Object Detection文章目录Cascaded Partial Decoder for Fast and Accurate Salient Object Detection几个亮点网络结构双分支训练整体注意力模块(Holistic Attention Module)解码器实验细...原创 2019-05-01 23:36:28 · 6349 阅读 · 33 评论 -
显著性检测之Pyramid Feature Attention Network for Saliency detection
Pyramid Feature Attention Network for Saliency detection文章目录Pyramid Feature Attention Network for Saliency detection主要结构CPFE:context-aware pyramid feature extractionChannel-wise attention & Spaci...原创 2019-05-04 15:53:06 · 4452 阅读 · 6 评论 -
显著性检测之BASNet: Boundary Aware Salient Object Detection
BASNet: Boundary Aware Salient Object Detection文章目录BASNet: Boundary Aware Salient Object Detection网络结构主要亮点损失函数实验细节相关链接原始文档:https://www.yuque.com/lart/papers网络结构主要亮点深监督编解码器(sup1~sup8,可见前图)额外...原创 2019-05-27 15:51:29 · 6001 阅读 · 0 评论 -
显著性检测之A Simple Pooling-Based Design for Real-Time Salient Object Detection
A Simple Pooling-Based Design for Real-Time Salient Object Detection文章目录A Simple Pooling-Based Design for Real-Time Salient Object Detection网络结构主要亮点大量的池化操作(FAM、GGM)使用边缘检测联合训练实验细节各模块的有效性联合训练的有效性效果比较相关...原创 2019-05-27 16:00:59 · 2027 阅读 · 0 评论 -
显著性检测之Attentive Feedback Network for Boundary-Aware Salient Object Detection
Attentive Feedback Network for Boundary-Aware Salient Object Detection文章目录Attentive Feedback Network for Boundary-Aware Salient Object Detection网络结构主要亮点明确的目标注意力反馈模块全局感知模块损失函数实验细节对比其他方法全局感知模块的配置比较模块消融...原创 2019-05-30 20:59:42 · 3985 阅读 · 18 评论 -
显著性检测之A Mutual Learning Method for Salient Object Detection with intertwined Multi-Supervision
A Mutual Learning Method for Salient Object Detection with intertwined Multi-Supervision文章目录A Mutual Learning Method for Salient Object Detection with intertwined Multi-Supervision主要亮点网络结构主要工作Mutual ...原创 2019-06-12 18:02:34 · 2829 阅读 · 3 评论 -
显著性检测之An Iterative and Cooperative Top-down and Bottom-up Inference Network for Salient Object Detec
An Iterative and Cooperative Top-down and Bottom-up Inference Network for Salient Object Detection文章目录An Iterative and Cooperative Top-down and Bottom-up Inference Network for Salient Object Detectio...原创 2019-06-12 18:04:46 · 2522 阅读 · 0 评论 -
显著性检测之Salient Object Detection with Pyramid Attention and Salient Edges
Salient Object Detection with Pyramid Attention and Salient Edges文章目录Salient Object Detection with Pyramid Attention and Salient Edges主要亮点网络结构Pyramid Attention Module(2c)Salient Edge Detector(2e)Dens...原创 2019-06-12 18:06:48 · 3228 阅读 · 6 评论 -
关于阅读论文的一些感想
关于阅读论文的一些感想最近看了一些论文,也和同学在交谈中学习到了一些关于阅读时应该有的态度,这里记录一下。原文地址:https://www.yuque.com/lart/papers/dgrl#EpF76关于实验部分,受到 语雀用户[@xiaoqi skr~skr(xiaoqizhao)] 的指导,在阅读论文时应该关注这样几个点:不要被作者的描述带偏,阅读的时候要留心,作者文字给你的...原创 2019-07-09 12:16:15 · 2895 阅读 · 0 评论 -
显著性检测之Detect Globally, Refine Locally: A Novel Approach to Saliency Detection
Detect Globally, Refine Locally: A Novel Approach to Saliency Detection文章目录Detect Globally, Refine Locally: A Novel Approach to Saliency DetectionRecurrent Localization Network (RLN)Contextual Weight...原创 2019-07-09 21:01:18 · 1469 阅读 · 2 评论 -
显著性检测之SE2Net: Siamese Edge-Enhancement Network for Salient Object Detection
SE2Net: Siamese Edge-Enhancement Network for Salient Object Detection文章目录SE2Net: Siamese Edge-Enhancement Network for Salient Object Detection网络结构主要贡献迭代过程边缘引导实验细节消融实验和其他模型比较一点感想参考链接原始文档:https://ww...原创 2019-07-04 15:45:01 · 2150 阅读 · 5 评论 -
显著性检测之Saliency Detection with Recurrent Fully Convolutional Networks
Saliency Detection with Recurrent Fully Convolutional Networkshttps://www.yuque.com/lart/papers现存的深度方法主要有以下一些问题.过去证明很有效的显著性先验被完全抛弃CNN的预测只考虑有限的局部图像区域. 不能加强空间一致性.使用前向传播架构, CNN很难细化输出的预测.显著性检测多被...原创 2019-03-02 16:25:29 · 2300 阅读 · 4 评论