Xuan Dong

Xuan Dong

Auto-generated profile, based on publicly available information

Xuan Dong

School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, China

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 March 2021 in IEEE Transactions on Affective Computing
Verified
Journal Article
Closed access
WLWeixin Li
Weixin Li
XDXuan Dong
Xuan Dong
YWyunhong wang
yunhong wang
Recognizing the emotional state of a person within the image in real-world scenarios is a key problem in affective computing and has various promising applications. Local regions in the image, including different objects in the background scene and parts within the foreground body, usually have different contributions to emotion perception of the target person. This, however, has not been well exploited in most existing methods. In this paper, we propose to make relational region-level analysis to account for the different contributions of different regions to emotion recognition. For the background scene, we propose a Body-Object Attention (BOA) module to estimate the contributions of background objects to emotion recognition given the target foreground body. Within the foreground body, we propose a Body Part Attention (BPA) module to recalibrate the channel-wise body feature responses to attend on body parts that are more important. Moreover, we propose to model the emotion label dependency in real-world images, considering both the semantic meanings of these labels and their co-occurrence patterns. We evaluate the proposed method on the EMOTIC and CAER-S datasets, and experimental results show the superiority of our method compared with the state-of-the-art algorithms.

25 February 2023 in ACM Transactions on Multimedia Computing, Communications, and Applications
Verified
Journal Article
Open access
KXKang Xu
Kang Xu
WLWeixin Li
Weixin Li
XWXia Wang
Xia Wang
XHXiaoyan Hu
Xiaoyan Hu
KYKe Yan
Ke Yan
XDXuan Dong
Xuan Dong
Image denoising is a fundamental problem in computer vision and multimedia computation. Non-local filters are effective for image denoising. But existing deep learning methods that use non-local computation structures are mostly designed for high-level tasks, and global self-attention is usually adopted. For the task of image denoising, they have high computational complexity and have a lot of redundant computation of uncorrelated pixels. To solve this problem and combine the marvelous advantages of non-local filter and deep learning, we propose a Convolutional Unbiased Regional (CUR) transformer. Based on the prior that, for each pixel, its similar pixels are usually spatially close, our insights are that (1) we partition the image into non-overlapped windows and perform regional self-attention to reduce the search range of each pixel, and (2) we encourage pixels across different windows to communicate with each other. Based on our insights, the CUR transformer is cascaded by a series of convolutional regional self-attention (CRSA) blocks with U-style short connections. In each CRSA block, we use convolutional layers to extract the query, key, and value features, namely Q , K , and V , of the input feature. Then, we partition the Q , K , and V features into local non-overlapped windows and perform regional self-attention within each window to obtain the output feature of this CRSA block. Among different CRSA blocks, we perform the unbiased window partition by changing the partition positions of the windows. Experimental results show that the CUR transformer outperforms the state-of-the-art methods significantly on four low-level vision tasks, including real and synthetic image denoising, JPEG compression artifact reduction, and low-light image enhancement.

8 September 2020 in IEEE Transactions on Visualization and Computer Graphics
Verified
Journal Article
Closed access
XDXuan Dong
Xuan Dong
WLWeixin Li
Weixin Li
XHXiaoyan Hu
Xiaoyan Hu
XWXiaojie Wang
Xiaojie Wang
YWyunhong wang
yunhong wang
In monochrome-color dual-lens systems, the monochrome camera can capture images with higher quality than the color camera. To obtain high quality color images, a better approach is to colorize the gray images from the monochrome camera with the color images from the color camera serving as a reference. In addition, the colorization may fail in some cases, which makes the estimation of the colorization quality a necessary step before outputting the colorization result. To solve these problems, we propose a deep convolutional network based framework. 1) In the colorization module, the proposed colorization CNN uses deep feature representations, attention operation, 3-D regulation and color correction to make use of colors of multiple pixels in the reference image for colorizing each pixel in the input gray image. 2) In the colorization quality estimation module, based on the symmetry property of colorization, we propose to utilize the colorization CNN again to colorize the gray map of the original reference color image using the first-time colorization result from the colorization module as reference. Then, the quality loss of the second-time colorization result can be used for estimating the colorization quality. Experimental results show that our method can largely outperform the state-of-the-art colorization methods and estimate the colorization quality accurately as well.

16 July 2021 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
XDXuan Dong
Xuan Dong
CLChang Liu
Chang Liu
WLWeixin Li
Weixin Li
XHXiaoyan Hu
Xiaoyan Hu
XWXiaojie Wang
Xiaojie Wang
YWyunhong wang
yunhong wang
Colorization in monochrome-color camera systems aims to colorize the gray image I G from the monochrome camera using the color image R C from the color camera as reference. Since monochrome cameras have better imaging quality than color cameras, the colorization can help obtain higher quality color images. Related learning based methods usually simulate the monochrome-color camera systems to generate the synthesized data for training, due to the lack of ground-truth color information of the gray image in the real data. However, the methods that are trained relying on the synthesized data may get poor results when colorizing real data, because the synthesized data may deviate from the real data. We present a self-supervised CNN model, named Cycle CNN, which can directly use the real data from monochrome-color camera systems for training. In detail, we use the Weighted Average Colorization (WAC) network to do the colorization twice. First, we colorize I G using R C as reference to obtain the first-time colorization result I C . Second, we colorize the de-colored map of R C , i.e. R G , using the concatenated image of I G and Cb/Cr channels of the first-time colorization result I C , i.e. I C Cb and I C Cr , as reference to obtain the second-time colorization result R C ' . In this way, for the second-time colorization result R C ' , we use the Cb and Cr channels of the original color map R C as ground-truth and introduce the cycle consistency loss to push R C 'Cb/Cr ≈ R C Cb/Cr . Also, for the Y channel of the first-time colorization result I C Y , we propose the Global Curve Adjustment (GCA) network and the structure similarity loss to encourage the structure similarity between I C Y and I G . In addition, we introduce a spatial smoothness loss within the WAC network to encourage spatial smoothness of the colorization result. Combining all these losses, we could train the Cycle CNN using the real data in the absence of the ground-truth color information of I G . Experimental results show that we can outperform related methods largely for colorizing real data.

1 April 2011
Verified
Conference Paper
Closed access
XBXiaole Bai
Xiaole Bai
ZYZiqiu Yun
Ziqiu Yun
DXDong Xuan
Dong Xuan
BCBiao Chen
Biao Chen
WZWei Zhao
Wei Zhao
Notice of Violation of IEEE Publication Principles "Optimal Multiple-coverage of Sensor Networks" by Xiaole Bai, Ziqiu Yun, Dong Xuan, Biao Chen and Wei Zhao, in the Proceedings of INFOCOM, pp.2498-2506, April 2011 After careful and considered review of the content and authorship of this paper by a duly constituted expert committee, this paper has been found to be in violation of IEEE's Publication Principles. This paper contains significant portions of original text from the paper cited below. The original text was copied without attribution (including appropriate references to the original author(s) and/or paper title) and without permission. One author (Ziqiu Yun) was responsible for the violation, and this violation was done without the knowledge of the other authors. "Planar Thinnest Deployment Pattern of Congruent Discs which Achieves 2-coverage" by Ge Jun, in his thesis presented at the School of Mathematical Sciences, Soochow University, Suzhou 215006, P. R. China, June 2010. In wireless sensor networks, multiple-coverage, in which each point is covered by more than one sensor, is often required to improve detection quality and achieve high fault tolerance. However, finding optimal patterns that achieve multiple-coverage in a plane remains a long-lasting open problem. In this paper, we first derive the optimal deployment density bound for two-coverage deployment patterns where Voronoi polygons generated by sensor nodes are congruent.We then propose optimal two-coverage patterns based on the optimal bound. We further extend these patterns by considering the connectivity requirement and design a set of optimal patterns that achieve two-coverage and one-, two-, and three-connectivity. We also study optimal patterns under practical considerations. To our knowledge, our work is the very first that proves the optimality of multiple-coverage deployment patterns.

25 October 2022 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
XDXuan Dong
Xuan Dong
CLChang Liu
Chang Liu
XHXiaoyan Hu
Xiaoyan Hu
KXKang Xu
Kang Xu
WLWeixin Li
Weixin Li
We study the colorization problem in monochrome-color dual-lens camera systems, i.e. colorizing the gray image from the monochrome camera using the color image from the color camera as reference. In related methods, cost volume based CNN methods achieve the state-of-the-art results, but they are costly in GPU memory due to building the 4D cost volume. Recently, some slice-wise cross-attention based methods are proposed for related problems. The slice-wise cross-attention has much less costs in GPU memory but directly using them for this colorization problem cannot generate competing results. We make use of the non-local computation property of cross-attention to propose a transformer based method. To overcome the limitations of straight-forward slice-wise cross-attention, we propose the spatially consistent cross-attention (SCCA) block to encourage pixels of slices across different epipolar lines in the gray image to find spatially consistent correspondence with pixels of the reference color image. And, to further reduce the memory cost while keeping the colorization accuracy, we design a pyramid processing strategy to cascade a series of SCCA blocks with smaller slice size and perform the colorization from coarse to fine. To extract more powerful image features, we use several regional self-attention (RSA) blocks with U-style connections. Experimental results show that we outperform the state-of-the-art methods largely on the synthesized datasets of Cityscapes, Sintel, and SceneFlow, and the real monochrome-color dual-lens dataset.