检索结果-内蒙古大学图书馆

3-D Context Entropy Model for Improved Practical Image Compression

学校读者我要写书评

暂无评论

3-D Context Entropy Model for Improved Practical Image Compr...

IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

作者： Zongyu Guo Yaojun Wu Runsen Feng Zhizheng Zhang Zhibo Chen CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China

ISBN: (数字)9781728193601

ISBN: (纸本)9781728193618

In this paper, we present our image compression framework designed for CLIC 2020 competition. Our method is based on Variational AutoEncoder (VAE) architecture which is strengthened with residual structures. In short, we make three noteworthy improvements here. First, we propose a 3-D context entropy model which can take advantage of known latent representation in current spatial locations for better entropy estimation. Second, a light-weighted residual structure is adopted for feature learning during entropy estimation. Finally, an effective training strategy is introduced for practical adaptation with different resolutions. Experiment results indicate our image compression method achieves 0.9775 MS-SSIM on CLIC validation set and 0.9809 MS-SSIM on test set.

关键词： Context modeling Solid modeling Image coding Entropy Training Transforms Image resolution

Learned Video Compression with Feature-level Residuals

学校读者我要写书评

暂无评论

Learned Video Compression with Feature-level Residuals

IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

作者： Runsen Feng Yaojun Wu Zongyu Guo Zhizheng Zhang Zhibo Chen CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China

ISBN: (数字)9781728193601

ISBN: (纸本)9781728193618

In this paper, we present an end-to-end video compression network for P-frame challenge on CLIC. We focus on deep neural network (DNN) based video compression, and improve the current frameworks from three aspects. First, we notice that pixel space residuals is sensitive to the prediction errors of optical flow based motion compensation. To suppress the relative influence, we propose to compress the residuals of image feature rather than the residuals of image pixels. Furthermore, we combine the advantages of both pixel-level and feature-level residual compression methods by model ensembling. Finally, we propose a step-by-step training strategy to improve the training efficiency of the whole framework. Experiment results indicate that our proposed method achieves 0.9968 MS-SSIM on CLIC validation set and 0.9967 MS-SSIM on test set.

关键词： Image coding Training Optical imaging Adaptive optics Motion compensation Optical distortion Decoding

Learned video compression with feature-level residuals

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Feng, Runsen Wu, Yaojun Guo, Zongyu Zhang, Zhizheng Jin, Xin Chen, Zhibo CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China

关键词： Deep neural networks

Multi-scale grouped dense network for VVC intra coding

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Li, Xin Sun, Simeng Zhang, Zhizheng Chen, Zhibo CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China

Versatile Video Coding (H.266/VVC) standard achieves better image quality when keeping the same bits than any other conventional image codec, such as BPG, JPEG, and etc. However, it is still attractive and challenging to improve the image quality with high compression ratio on the basis of traditional coding techniques. In this paper, we design the multi-scale grouped dense network (MSGDN) to further reduce the compression artifacts by combining the multi-scale and grouped dense block, which are integrated as the post-process network of VVC intra coding. Besides, to improve the subjective quality of compressed image, we also present a generative adversarial network (MSGDN-GAN) by utilizing our MSGDN as generator. Across the extensive experiments on validation set, our MSGDN trained by MSE losses yields the PSNR of 32.622 on average with teams"IMC" and"haha" at the bit-rate of 0.15 in Low-rate track. Moreover, our MSGDN-GAN could achieve the better subjective performance. Copyright © 2020, The Authors. All rights reserved.

关键词： Image quality

From multi-view to hollow-3D: Hallucinated hollow-3D R-CNN for 3D object detection

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Deng, Jiajun Zhou, Wengang Zhang, Yanyong Li, Houqiang The CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei230026 China The Department of Computer Science University of Science and Technology of China Hefei230026 China

As an emerging data modal with precise distance sensing, LiDAR point clouds have been placed great expectations on 3D scene understanding. However, point clouds are always sparsely distributed in the 3D space, and with unstructured storage, which makes it difficult to represent them for effective 3D object detection. To this end, in this work, we regard point clouds as hollow-3D data and propose a new architecture, namely Hallucinated Hollow-3D R-CNN (H23D R-CNN), to address the problem of 3D object detection. In our approach, we first extract the multi-view features by sequentially projecting the point clouds into the perspective view and the bird-eye view. Then, we hallucinate the 3D representation by a novel bilaterally guided multi-view fusion block. Finally, the 3D objects are detected via a box refinement module with a novel Hierarchical Voxel RoI Pooling operation. The proposed H23D R-CNN provides a new angle to take full advantage of complementary information in the perspective view and the bird-eye view with an efficient framework. We evaluate our approach on the public KITTI Dataset and Waymo Open Dataset. Extensive experiments demonstrate the superiority of our method over the state-of-the-art algorithms with respect to both effectiveness and efficiency. The code will be made available at https://***/djiajunustc/H-23D_R-CNN. Copyright © 2021, The Authors. All rights reserved.

关键词： Object recognition

学校读者我要写书评

暂无评论

IEEE International Conference on Multimedia and Expo (ICME)

作者： Xiaoke Zhu Danyang Li Xiaopan Chen Fumin Qi Fan Zhang Xiao-Yuan Jing School of Computer and Information Engineering Henan University Kaifeng China Henan Engineering Research Center of Intelligent Technology and Application Henan University Kaifeng China Department of System Operation National Supercomuting Center in Shenzhen Shenzhen China Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng China School of Computer Science Wuhan University Wuhan China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Facial image based kinship verification aims to decide whether there exists kinship between the given facial images. In practice, the cross-generation differences will cause adverse effects on kinship verification, which limits the performance. Therefore, how to mine the implied similarity from facial images with large cross-generation divergence is an important problem in kinship verification, which has not yet been well studied. In view of this, we propose a Similarity Mining via Implicit matching pattern LEarning (SMILE) approach for kinship verification. Specifically, SMILE mainly consists of two modules, including a Semi-coupled Multi-pattern Similarity Learning (SMSL) module and a Cross-Generation Feature Normalization (CGFN) module. The SMSL module is designed to learn multiple semi-coupled matching patterns for mining the implicit facial similarity information from different perspectives. The CGFN module aims to reduce the divergence between facial images of parent and child. Extensive experiments demonstrate that the proposed approach outperforms the existing state-of-the-art methods.

关键词： Pattern matching Similarity learning

spatial-temporal water area monitoring of Miyun Reservoir using remote sensing imagery from 1984 to 2020

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Chang Tang, Hairong Ji, Luyan Zhao, Yongchao Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China University of Chinese Academy of Sciences Beijing100049 China Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Beijing100190 China

Miyun Reservoir has produced huge benefits in flood control, agricultural irrigation, power generation, aquaculture, tourism, and urban water supply. Accurately water mapping is of great significance to the ecological environment monitoring of the Miyun Reservoir and the management of the South-to-North Water Diversion Project. On the 60th anniversary of the completion of the Miyun Reservoir, we took the Miyun Reservoir as the study area and collected all the Landsat-5 and Landsat-8 remote sensing images from 1984 to 2020 for water mapping. Based on the spectral, topographical and temporal-spatial characteristics of water, we proposed an automated method for long-term researvoir mapping, which can solve the problems caused by cloud, shadow, ice and snow pixels. Moreover, it can also deal with the "same objects with different spectra" and spectral mixed problems. The overall accuracy is as high as 98.2% for the case with no cloud or snow/ice cover. The landscape division index is introduced to analyze the morphological changes of Miyun Reservoir. Based on the mapping results, it is suggested that the water area change of the reservoir can be divided into five periods: "growth period" (1984-1993), "peak period" (1994-1999), "decline period" (2000-2003), "protection period" "(2004-2014), and "recovery period" (2015-2020), during which the east and west areas of the reservoir have experienced the process of "division-combination-division-combination". The largest reservoir area reached 151.6 km2 (1995), and the smallest area was only 57.3 km2 (2004). Moreover, we also find that the largest area in a year occurs from August to September, and the smallest area occurs in May. In addition, we have found that the changes of water area of Miyun Reservoir are closely connected to urban development, government policies and precipitation varitations. The study shows that our method used here are potentially extendable to other reservoirs for long-term area monitoring. © 2021, CC

关键词： Time series analysis

Texture-enhanced Light Field Super-resolution with Spatio-Angular Decomposition Kernels

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Hu, Zexi Chen, Xiaoming Wing, Henry Yeung, Fung Chung, Yuk Ying Chen, Zhibo School of Computer Science University of Sydney Australia School of Computer Science and Engineering Beijing Technology and Business University China CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China China

Despite the recent progress in light field super-resolution (LFSR) achieved by convolutional neural networks, the correlation information of light field (LF) images has not been sufficiently studied and exploited due to the complexity of 4D LF data. To cope with such high-dimensional LF data, most of the existing LFSR methods resorted to decomposing it into lower dimensions and subsequently performing optimization on the decomposed sub-spaces. However, these methods are inherently limited as they neglected the characteristics of the decomposition operations and only utilized a limited set of LF sub-spaces ending up failing to sufficiently extract spatio-angular features and leading to a performance bottleneck. To overcome these limitations, in this paper, we comprehensively discover the potentials of LF decomposition and propose a novel concept of decomposition kernels. In particular, we systematically unify the decomposition operations of various sub-spaces into a series of such decomposition kernels, which are incorporated into our proposed Decomposition Kernel Network (DKNet) for comprehensive spatio-angular feature extraction. The proposed DKNet is experimentally verified to achieve considerable improvements compared with the state-of-the-art methods. To further improve DKNet in producing more visually pleasing LFSR results, based on the VGG network, we propose a LFVGG loss to guide the Texture-Enhanced DKNet (TE-DKNet) to generate rich authentic textures and enhance LF images’ visual quality significantly. We also propose an indirect evaluation metric by taking advantage of LF material recognition to objectively assess the perceptual enhancement brought by the LFVGG loss. Copyright © 2021, The Authors. All rights reserved.

关键词： Optical resolving power

A Fusion processing Method for Satellite Detection Data by Beidou Short Message system 12th

学校读者我要写书评

暂无评论

A Fusion Processing Method for Satellite Detection Data by B...

12th China Satellite Navigation Conference, CSNC 2021

作者： Li, Bing Yu, Xiao Sun, Xiaojuan Hu, Yuxin Wu, Haiyan Shi, Tao Chen, Ling School of Cyber Security University of Chinese Academy of Sciences Beijing100149 China State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China Aerospace Information Research Institute Chinese Academy of Sciences Beijing100190 China Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Chinese Academy of Sciences Beijing100190 China Beijing Institute of Tracking and Telecommunications Technology Beijing100094 China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing100149 China National Space Science Center Chinese Academy of Science Beijing100190 China

ISBN: (纸本)9789811631375

With the support of Beidou terminal of short message service system, the space science satellite can transmit all-day data that are astronomical alert data such as gravitational waves and Gamma ray bursts and satellite telemetry parameters. Then Beidou message service system has become an important means for finding the space astronomical events quickly, planning the follow-up observation and monitoring the real-time health of satellite and payloads. Then the application fields of the Beidou system has expanded into space science observation and satellite emergency operation. The short message data receiving and processing in ground system has many features of near real-time and continuity which is different from the processing methods of satellite data transmitted by data channel and telemetry channel. And the multi-source satellite data of the telemetry parameters and the astronomical alert information from these three channels need to be integrated. So, the real-time processing system of satellite data based on Beidou short message transmission has great challenges. In this paper, we explored an organization method of Beidou short messages, and proposed a multi-source data fusion algorithm for the ground satellite data processing system. The data processing system with this method was designed and implemented. Applying to the ground system of the satellite named gravitational wave high-energy electromagnetic counterpart all-sky monitor, the performance analysis of improved system shows that the processing delay for important telemetry parameters and astronomical alert information is greatly shortened from average nine hours to several minutes. The Beidou short message service system shows great supports in real time observation of gamma ray bursts. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Satellites