基于注意力机制与图卷积神经网络的单目红外图像深度估计
Depth estimation of monocular infrared images based on attention mechanism and graph convolutional neural network
查看参考文献17篇
|
文摘
|
对场景中的物体进行深度估计是无人驾驶领域中的关键问题,红外图像有利于在光线不佳的情况下解决深度估计问题。针对红外图像纹理不清晰与边缘信息不丰富的特点,提出了将注意力机制与图卷积神经网络相结合来解决单目红外图像深度估计问题。首先,在深度估计问题中,图像中每个像素点的深度信息不仅与其周围像素点的深度信息相关,还需考虑更大范围的其他像素点的深度信息,采用注意力机制可以针对这一点有效提取图像的像素级全局深度信息关联。其次,基于深度信息关联得到的特征可以考虑为非欧数据,进一步使用图卷积神经网络(graph convolutional neural network, GCN)来进行推理。最后,在训练阶段将连续的深度估计回归问题转化成分类问题,使训练过程更稳定,降低了网络的学习难度。实验结果表明,该方法在红外数据集NUST-SR上获得了良好的效果,在阈值指标小于1.253时,准确率提升了1.2%,相较其他方法更具优势。 |
|
其他语种文摘
|
The depth estimation of objects in the scene is a key issue in the field of the unmanned driving. The infrared images are helpful to solve the depth estimation problem under poor light conditions. Aiming at characteristics of unclear infrared images texture and insufficient edge information, a combination of attention mechanism and graph convolutional neural network was proposed to solve the problem of monocular infrared images depth estimation. First of all, in the depth estimation problem, the depth information of each pixel in the image was not only related to the depth information of its surrounding pixels, but also needed to consider the depth information of a larger range of other pixels. The attention mechanism could be effectively extract the pixel-level global depth information association of images. Secondly, the features obtained based on the depth information association could be considered as non-Euclidean data, and the graph convolutional neural network (GCN) was further used for reasoning. Finally, in the training phase, the continuous depth estimation regression problem was transformed into the classification problem, which made the training process more stable and reduced the learning difficulty of the network. The experimental results show that the proposed method has obtained good results on the infrared data set NUST-SR. When the threshold index is less than 1.253, the accuracy rate is improved by 1.2%, which is more advantageous than other methods. |
|
来源
|
应用光学
,2021,42(1):49-56 【扩展库】
|
|
DOI
|
10.5768/jao202142.0102001
|
|
关键词
|
红外图像
;
深度估计
;
注意力机制
;
图卷积神经网络
|
|
地址
|
华东理工大学信息科学与工程学院, 上海, 200237
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
1002-2082 |
|
学科
|
电子技术、通信技术;自动化技术、计算机技术 |
|
基金
|
国家自然科学基金面上项目
|
|
文献收藏号
|
CSCD:6919031
|
参考文献 共
17
共1页
|
|
1.
Silberman N. Indoor segmentation and support inference from RGBD Images.
Proceedings of the 12th European conference on Computer Vision,2012:740-746
|
CSCD被引
1
次
|
|
|
|
|
2.
Eigen D. Depth map prediction from a single image using a multi-scale deep network.
NIPS'14 Proceedings of the 27th International Conference on Neural Information Processing Systems,2014:2366-2374
|
CSCD被引
2
次
|
|
|
|
|
3.
Eigen D. Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture.
2015 IEEE International Conference on Computer Vision (ICCV),2015:2650-2658
|
CSCD被引
7
次
|
|
|
|
|
4.
Laina I. Deeper depth prediction with fully convolutional residual networks.
2016 Fourth International Conference on 3D Vision (3DV),2016:239-248
|
CSCD被引
13
次
|
|
|
|
|
5.
He Kaiming. Deep residual learning for image recognition.
2016 IEEE CVF Conference on Computer Vision and Pattern Recognition,2016:770-778
|
CSCD被引
1
次
|
|
|
|
|
6.
吴寿川. 基于双向递归卷积神经网络的单目红外视频深度估计.
光学学报,2017,37(12):246-254
|
CSCD被引
2
次
|
|
|
|
|
7.
顾婷婷. 基于帧间信息提取的单幅红外图像深度估计.
激光与光电子学进展,2018,55(6):163-172
|
CSCD被引
2
次
|
|
|
|
|
8.
Fisher Y. Multi-scale context aggregation by dilated convolutions.
arXiv,2016: 1511.07122
|
CSCD被引
1
次
|
|
|
|
|
9.
Li Bo. Monocular depth estimation with hierarchical fusion of dilated CNNs and soft-weighted-sum inference.
Pattern Recognition,2018,83:328-339
|
CSCD被引
14
次
|
|
|
|
|
10.
Fu H. Deep ordinal regression network for monocular depth estimation.
2018 IEEE CVF Conference on Computer Vision and Pattern Recognition,2018:2002-2011
|
CSCD被引
2
次
|
|
|
|
|
11.
Bahdanau D. Neural machine translation by jointly learning to align and translate.
arXiv,2015: 1409.0473
|
CSCD被引
1
次
|
|
|
|
|
12.
Xu Dan. Structured attention guided convolutional neural fields for monocular depth estimation.
2018 IEEE CVF Conference on Computer Vision and Pattern Recognition,2018:3917-3925
|
CSCD被引
1
次
|
|
|
|
|
13.
Li R B. Deep attention-based classification network for robust depth prediction.
Computer Vision-ACCV 2018,2019:663-678
|
CSCD被引
1
次
|
|
|
|
|
14.
陈裕如. 基于自适应像素级注意力模型的场景深度估计.
应用光学,2020,41(3):490-499
|
CSCD被引
2
次
|
|
|
|
|
15.
Fu Junwei. Monocular depth estimation based on multi-scale graph convolution networks.
IEEE Access,2020(8):997-1009
|
CSCD被引
4
次
|
|
|
|
|
16.
Xu Keyulu. Representation learning on graphs with jumping knowledge networks.
arXiv,2018: 1806.03536
|
CSCD被引
1
次
|
|
|
|
|
17.
Simonyan K.
Very deep convolutional networks for large-scale image recognition,2014:1409-1556
|
CSCD被引
5
次
|
|
|
|
|
了解气象卫星更多信息，请访问