论文阅读 [TPAMI-2022] DeFusionNET: Defocus Blur Detection via Recurrently Fusing and Refining Discrimi

论文阅读 [TPAMI-2022] DeFusionNET: Defocus Blur Detection via Recurrently Fusing and Refining Discriminative Multi-Scale Deep Features

论文搜索(studyai.com)

关键字(Keywords)

Feature extraction; Neural networks; Semantics; Image edge detection; Fuses; Task analysis; Machine learning; Defocus blur detection; multi-scale features; feature fusing; channel attention

机器视觉; 自然语言处理

语义分析; 语义特征; 模糊检测; 边缘检测

摘要(Abstract)

Albeit great success has been achieved in image defocus blur detection, there are still several unsolved challenges, e.g., interference of background clutter, scale sensitivity and missing boundary details of blur regions.

尽管在图像离焦模糊检测方面取得了巨大的成功，但仍然存在一些尚未解决的挑战，例如背景杂波干扰、尺度敏感性和模糊区域边界细节缺失。.

To deal with these issues, we propose a deep neural network which recurrently fuses and refines multi-scale deep features (DeFusionNet) for defocus blur detection.

为了解决这些问题，我们提出了一种深度神经网络，该网络反复融合和细化多尺度深度特征（defusinet），用于散焦模糊检测。.

We first fuse the features from different layers of FCN as shallow features and semantic features, respectively.

我们首先将FCN不同层次的特征分别融合为浅层特征和语义特征。.

Then, the fused shallow features are propagated to deep layers for refining the details of detected defocus blur regions, and the fused semantic features are propagated to shallow layers to assist in better locating blur regions.

然后，将融合的浅层特征传播到深层以细化检测到的散焦模糊区域的细节，并将融合的语义特征传播到浅层以帮助更好地定位模糊区域。.

The fusion and refinement are carried out recurrently.

融合和细化是反复进行的。.

In order to narrow the gap between low-level and high-level features, we embed a feature adaptation module before feature propagating to exploit the complementary information as well as reduce the contradictory response of different feature layers.

为了缩小低层和高层特征之间的差距，我们在特征传播之前嵌入了一个特征自适应模块，以利用互补信息，同时减少不同特征层之间的矛盾响应。.

Since different feature channels are with different extents of discrimination for detecting blur regions, we design a channel attention module to select discriminative features for feature refinement.

由于不同的特征通道在检测模糊区域时具有不同的区分程度，我们设计了通道注意模块来选择区分性特征进行特征细化。.

Finally, the output of each layer at last recurrent step are fused to obtain the final result.

最后，对每一层在最后一步的输出进行融合，得到最终结果。.

We collect a new dataset consists of various challenging images and their pixel-wise annotations for promoting further study.

我们收集了一个新的数据集，包括各种具有挑战性的图像及其像素级注释，以促进进一步的研究。.

Extensive experiments on two commonly used datasets and our newly collected one are conducted to demonstrate both the efficacy and efficiency of DeFusionNet…

在两个常用数据集和我们新收集的数据集上进行了广泛的实验，以证明去氟的有效性和效率。。.

作者(Authors)

[‘Chang Tang’, ‘Xinwang Liu’, ‘Xiao Zheng’, ‘Wanqing Li’, ‘Jian Xiong’, ‘Lizhe Wang’, ‘Albert Y. Zomaya’, ‘Antonella Longo’]

本文来自互联网用户投稿，文章观点仅代表作者本人，不代表本站立场，不承担相关法律责任。如若转载，请注明出处。 如若内容造成侵权/违法违规/事实不符，请点击【内容举报】进行投诉反馈！