AI帮你理解科学

AI 生成解读视频

AI抽取解析论文重点内容自动生成视频


pub
生成解读视频

AI 溯源

AI解析本论文相关学术脉络


Master Reading Tree
生成 溯源树

AI 精读

AI抽取本论文的概要总结


微博一下
The experimental results on two standard benchmarks show that our proposed DeBLuRing Network and DeBLuRring Generative Adversarial Network outperform the existing state-of-the-art methods in video deblurring

Adversarial Spatio-Temporal Learning for Video Deblurring.

IEEE Transactions on Image Processing, no. 1 (2019): 291-301

被引用31|浏览176
EI WOS

摘要

Camera shake or target movement often leads to undesired blur effects in videos captured by a hand-held camera. Despite significant efforts having been devoted to video-deblur research, two major challenges remain: (1) how to model the spatio-temporal characteristics across both the spatial domain (i.e., image plane) and the temporal doma...更多

代码

数据

0
简介
  • Videos captured by hand-held cameras often suffer from unwanted blurs either caused by camera shake [1], or object movement in the scene [2], [3].
  • The task of video deblurring aims at removing those undesired blurs and recovering sharp frames from the input video.
  • This is an active research topic in the applied fields of computer vision and image processing.
  • The commonly adopted performance metric, namely, pixel-wise residual error, often measured by PSNR, is questionable, as it fails to capture human visual intuitions of how sharp or how realistic a restored image is [7], [8].
  • The authors plan to leverage the recent advance of the adversarial learning technique to improve the performance of video deblurring
重点内容
  • Videos captured by hand-held cameras often suffer from unwanted blurs either caused by camera shake [1], or object movement in the scene [2], [3]
  • Based on the above DeBLuRing Network, we develop a generative adversarial network, called DeBLuRring Generative Adversarial Network, with both content and adversarial losses
  • Inspired by the adversarial training strategy, we propose a model called DeBLuRring Generative Adversarial Network (DBLRGAN), which utilizes G to deblur images and D to discriminate deblurred images and real-world sharp images
  • We conduct experiments to demonstrate the effectiveness of the proposed DeBLuRing Network and DeBLuRring Generative Adversarial Network on the task of video deblurring
  • We demonstrated that DeBLuRing Network is able to capture better spatio-temporal features, leading to improved blur removal
  • The experimental results on two standard benchmarks show that our proposed DeBLuRing Network and DeBLuRring Generative Adversarial Network outperform the existing state-of-the-art methods in video deblurring
方法
  • INPUT PSDEBLUR

    WFA DBN DBN DBN DBLRNet DBLRNet DBLRNet DBLRGAN Average (PSNR)

    27.14 25.08 28.35 28.37 30.05 30.05 29.98 31.56 33.04 33.19 increases the number of training samples.
  • In DBLRGAN, the authors set the hyper parameter α as 0.0002 when the authors conduct experiments as empirically this value achieves the best performance
  • It has a better PSNR value due to three reasons.
  • The authors set the hyper parameter α as 0.0002 when the authors conduct experiments
  • This is a very small value, which forces the content loss to have an overwhelming superiority over the adversarial loss on PSNR value during the training stage.
  • Kim et al [50] evaluate DBN model and find that the speed of DBN model without aligning is almost more than 20 times faster than it with aligning because (a)
结果
  • The authors conduct experiments to demonstrate the effectiveness of the proposed DBLRNet and DBLRGAN on the task of video deblurring.
  • Su et al build a benchmark which contains videos captured by various kinds of devices such as iPhone 6s, GoPro Hero 4 and Nexus 5x, and each video includes about 100 frames of size 1280 × 720 [48].
  • This benchmark consists of two sub datasets: quantitative and qualitative ones.
  • Note that there is not ground truth for the qualitative subset, wSehacrapnor Blurry ?
结论
  • The authors have resorted to spatio-temporal learning and adversarial training to recover sharp and realistic video frames for video deblurring.
  • The authors proposed two novel network models.
  • The authors' second contribution is DBLRGAN equipped with both the content loss and adversarial loss, which are complementary to each other, driving the model to generate visually realistic images.
  • The experimental results on two standard benchmarks show that the proposed DBLRNet and DBLRGAN outperform the existing state-of-the-art methods in video deblurring
总结
  • Introduction:

    Videos captured by hand-held cameras often suffer from unwanted blurs either caused by camera shake [1], or object movement in the scene [2], [3].
  • The task of video deblurring aims at removing those undesired blurs and recovering sharp frames from the input video.
  • This is an active research topic in the applied fields of computer vision and image processing.
  • The commonly adopted performance metric, namely, pixel-wise residual error, often measured by PSNR, is questionable, as it fails to capture human visual intuitions of how sharp or how realistic a restored image is [7], [8].
  • The authors plan to leverage the recent advance of the adversarial learning technique to improve the performance of video deblurring
  • Methods:

    INPUT PSDEBLUR

    WFA DBN DBN DBN DBLRNet DBLRNet DBLRNet DBLRGAN Average (PSNR)

    27.14 25.08 28.35 28.37 30.05 30.05 29.98 31.56 33.04 33.19 increases the number of training samples.
  • In DBLRGAN, the authors set the hyper parameter α as 0.0002 when the authors conduct experiments as empirically this value achieves the best performance
  • It has a better PSNR value due to three reasons.
  • The authors set the hyper parameter α as 0.0002 when the authors conduct experiments
  • This is a very small value, which forces the content loss to have an overwhelming superiority over the adversarial loss on PSNR value during the training stage.
  • Kim et al [50] evaluate DBN model and find that the speed of DBN model without aligning is almost more than 20 times faster than it with aligning because (a)
  • Results:

    The authors conduct experiments to demonstrate the effectiveness of the proposed DBLRNet and DBLRGAN on the task of video deblurring.
  • Su et al build a benchmark which contains videos captured by various kinds of devices such as iPhone 6s, GoPro Hero 4 and Nexus 5x, and each video includes about 100 frames of size 1280 × 720 [48].
  • This benchmark consists of two sub datasets: quantitative and qualitative ones.
  • Note that there is not ground truth for the qualitative subset, wSehacrapnor Blurry ?
  • Conclusion:

    The authors have resorted to spatio-temporal learning and adversarial training to recover sharp and realistic video frames for video deblurring.
  • The authors proposed two novel network models.
  • The authors' second contribution is DBLRGAN equipped with both the content loss and adversarial loss, which are complementary to each other, driving the model to generate visually realistic images.
  • The experimental results on two standard benchmarks show that the proposed DBLRNet and DBLRGAN outperform the existing state-of-the-art methods in video deblurring
表格
  • Table1: CONFIGURATIONS OF THE PROPOSED DBLRNET. IT IS COMPOSED OF TWO CONVOLUTIONAL LAYERS (L1 AND L2), 14 RESIDUAL BLOCKS, TWO CONVOLUTIONAL LAYERS (L31 AND L32) WITHOUT SKIP CONNECTION, AND THREE ADDITIONAL CONVOLUTIONAL LAYERS (L33, L34 AND L35). EACH RESIDUAL BLOCK CONTAINS TWO CONVOLUTIONAL LAYERS, WHICH ARE INDICATED BY L(X) AND L(X+1) IN THE TABLE, WHERE “X”
  • Table2: CONFIGURATIONS OF OUR D MODEL IN DBLRGAN. BN MEANS BATCH NORMALIZATION AND RELU REPRESENTS THE ACTIVATION FUNCTION
  • Table3: PERFORMANCE COMPARISONS WITH [<a class="ref-link" id="c28" href="#r28">28</a>], [<a class="ref-link" id="c56" href="#r56">56</a>] AND [<a class="ref-link" id="c33" href="#r33">33</a>] ON THE BLURRED KITTI DATASET IN TERMS OF THE PSNR CRITERION. THE BEST RESULTS ARE SHOWN IN BOLD, AND THE SECOND BEST ARE
  • Table4: PERFORMANCE COMPARISONS IN TERMS PSNR WITH PSDEBLUR, WFA [<a class="ref-link" id="c32" href="#r32">32</a>], DBN (SINGLE), DBN (NOALIGN), DBN(FLOW) [<a class="ref-link" id="c48" href="#r48">48</a>], DBLRNET (SINGLE) AND DBLRNET (MULTI) ON THE VIDEODEBLURRING DATASET. THE BEST RESULTS ARE SHOWN IN BOLD, AND THE SECOND BEST ARE
Download tables as Excel
相关工作
  • Many approaches have been proposed for image/video deblurring, which can be roughly classified into two categories: geometry-based methods and deep learning methods.

    Geometry-based methods. Modern single-image deblurring methods iteratively estimate uniform or non-uniform blur kernels and the latent sharp image given a single blurry image [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20]. However, it is difficult for single image based methods to estimate kernel because blur is spatially varying in real world. To employ additional information, multi-image based methods [21], [22], [23], [24], [25], [26], [27] have been proposed to address blur, such as flash/no-flash image pairs [23], blurred/noise image pairs [22] and gyroscope information [24]. In order to accurately estimate kernels, some methods also use optical flow [28] and temporal information [29]. However, most of these methods are limited by the performance of an assumed degradation model and its estimation, thus some of them are fragile and cannot handle more challenging cases.
基金
  • Kaihao Zhang’s PhD scholarship is funded by Australian National University
  • Yiran Zhong’s PhD scholarship is funded by CSIRO Data61
  • Hongdong Li is CI (Chief Inivestigator) on Australia Centre of Excellence for Robotic Vision (CE14) funded by Australia Research Council
  • This work is also supported by 2017 Tencent Rhino Bird Elite Graduate Program
引用论文
  • S. B. Kang, “Automatic removal of chromatic aberration from a single image,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
    Google ScholarLocate open access versionFindings
  • J. Sun, W. Cao, Z. Xu, and J. Ponce, “Learning a convolutional neural network for non-uniform motion blur removal,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
    Google ScholarLocate open access versionFindings
  • J. Shi, L. Xu, and J. Jia, “Just noticeable defocus blur detection and estimation,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
    Google ScholarLocate open access versionFindings
  • H. Seok Lee and K. Mu Lee, “Dense 3d reconstruction from severely blurred images using a single moving camera,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2013.
    Google ScholarLocate open access versionFindings
  • H. S. Lee, J. Kwon, and K. M. Lee, “Simultaneous localization, mapping and deblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2011.
    Google ScholarLocate open access versionFindings
  • H. Jin, P. Favaro, and R. Cipolla, “Visual tracking in the presence of motion blur,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 18–25, 2005.
    Google ScholarLocate open access versionFindings
  • Z. Wang, E. P. Simoncelli, and A. C. Bovik, “Multiscale structural similarity for image quality assessment,” in Asilomar Conference on Signals, Systems and Computers (ACSSC), 2003.
    Google ScholarLocate open access versionFindings
  • Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600–612, 2004.
    Google ScholarLocate open access versionFindings
  • A. Gupta, N. Joshi, C. L. Zitnick, M. Cohen, and B. Curless, “Single image deblurring using motion density functions,” in European Conference on Computer Vision (ECCV), 2010.
    Google ScholarLocate open access versionFindings
  • M. Hirsch, C. J. Schuler, S. Harmeling, and B. Scholkopf, “Fast removal of non-uniform camera shake,” in The IEEE International Conference on Computer Vision (ICCV), 2011.
    Google ScholarLocate open access versionFindings
  • Z. Hu, L. Xu, and M.-H. Yang, “Joint depth estimation and camera shake removal from single blurry image,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014.
    Google ScholarLocate open access versionFindings
  • M. Jin, S. Roth, and P. Favaro, “Noise-blind image deblurring,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • P. P. Srinivasan, R. Ng, and R. Ramamoorthi, “Light field blind motion deblurring,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • Y. Yan, W. Ren, Y. Guo, R. Wang, and X. Cao, “Image deblurring via extreme channels prior,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • J. Pan, J. Dong, Y.-W. Tai, Z. Su, and M.-H. Yang, “Learning discriminative data fitting functions for blind image deblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • D. Gong, M. Tan, Y. Zhang, A. van den Hengel, and Q. Shi, “Selfpaced kernel estimation for robust blind image deblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • J. Dong, J. Pan, Z. Su, and M.-H. Yang, “Blind image deblurring with outlier handling,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • Y. Bahat, N. Efrat, and M. Irani, “Non-uniform blind deblurring by reblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • M. M. R. Mohan, A. N. Rajagopalan, and G. Seetharaman, “Going unconstrained with rolling shutter deblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • H. Park and K. Mu Lee, “Joint estimation of camera pose, depth, deblurring, and super-resolution from a blurred image sequence,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • A. Ito, A. C. Sankaranarayanan, A. Veeraraghavan, and R. G. Baraniuk, “Blurburst: Removing blur due to camera shake using multiple images,” ACM Transactions on Graphics (TOG), vol. 3, no. 1, 2014.
    Google ScholarLocate open access versionFindings
  • L. Yuan, J. Sun, L. Quan, and H.-Y. Shum, “Image deblurring with blurred/noisy image pairs,” in ACM Transactions on Graphics (TOG), vol. 26, p. 1, ACM, 2007.
    Google ScholarLocate open access versionFindings
  • G. Petschnigg, R. Szeliski, M. Agrawala, M. Cohen, H. Hoppe, and K. Toyama, “Digital photography with flash and no-flash image pairs,” in ACM Transactions on Graphics (TOG), vol. 23, pp. 664–672, ACM, 2004.
    Google ScholarLocate open access versionFindings
  • S. Hee Park and M. Levoy, “Gyro-based multi-image deconvolution for removing handshake blur,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014.
    Google ScholarLocate open access versionFindings
  • Y.-W. Tai, H. Du, M. S. Brown, and S. Lin, “Image/video deblurring using a hybrid camera,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2008.
    Google ScholarLocate open access versionFindings
  • P. Wieschollek, M. Hirsch, B. Scholkopf, and H. P. A. Lensch, “Learning blind motion deblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • W. Ren, J. Pan, X. Cao, and M.-H. Yang, “Video deblurring via semantic segmentation and pixel-wise non-linear kernel,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • T. Hyun Kim and K. Mu Lee, “Generalized video deblurring for dynamic scenes,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
    Google ScholarLocate open access versionFindings
  • Y. Li, S. B. Kang, N. Joshi, S. M. Seitz, and D. P. Huttenlocher, “Generating sharp panoramas from motion-blurred videos,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010.
    Google ScholarLocate open access versionFindings
  • N. M. Law, C. D. Mackay, and J. E. Baldwin, “Lucky imaging: high angular resolution imaging in the visible from the ground,” Astronomy & Astrophysics, vol. 446, no. 2, pp. 739–745, 2006.
    Google ScholarLocate open access versionFindings
  • S. Cho, J. Wang, and S. Lee, “Video deblurring for hand-held cameras using patch-based synthesis,” ACM Transactions on Graphics (TOG), vol. 31, no. 4, p. 64, 2012.
    Google ScholarLocate open access versionFindings
  • M. Delbracio and G. Sapiro, “Burst deblurring: Removing camera shake through fourier burst accumulation,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
    Google ScholarLocate open access versionFindings
  • L. Pan, Y. Dai, M. Liu, and F. Porikli, “Simultaneous stereo video deblurring and scene flow estimation,” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: Towards real-time object detection with region proposal networks,” in Advances in Neural Information Processing Systems (NIPS), 2015.
    Google ScholarLocate open access versionFindings
  • A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in Advances in Neural Information Processing Systems (NIPS), 2012.
    Google ScholarLocate open access versionFindings
  • K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 2015.
    Google ScholarFindings
  • C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, “Going deeper with convolutions,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
    Google ScholarLocate open access versionFindings
  • K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
    Google ScholarLocate open access versionFindings
  • Y. Sun, X. Wang, and X. Tang, “Deep convolutional network cascade for facial point detection,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
    Google ScholarLocate open access versionFindings
  • Y. Sun, X. Wang, and X. Tang, “Deep learning face representation from predicting 10,000 classes,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
    Google ScholarLocate open access versionFindings
  • K. Zhang, Y. Huang, C. Song, H. Wu, and L. Wang, “Kinship verification with deep convolutional neural networks,” in The British Machine Vision Conference (BMVC), 2015.
    Google ScholarLocate open access versionFindings
  • K. Zhang, Y. Huang, Y. Du, and L. Wang, “Facial expression recognition based on deep evolutional spatial-temporal networks,” IEEE Transactions on Image Processing, 2017.
    Google ScholarLocate open access versionFindings
  • S. Ji, W. Xu, M. Yang, and K. Yu, “3d convolutional neural networks for human action recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 1, pp. 221–231, 2013.
    Google ScholarLocate open access versionFindings
  • K. Simonyan and A. Zisserman, “Two-stream convolutional networks for action recognition in videos,” in Advances in Neural Information Processing Systems (NIPS), 2014.
    Google ScholarLocate open access versionFindings
  • C. Ledig, L. Theis, F. Huszar, J. Caballero, A. Cunningham, A. Acosta, A. Aitken, A. Tejani, J. Totz, Z. Wang, et al., “Photo-realistic single image super-resolution using a generative adversarial network,” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • L. Xu, J. S. Ren, C. Liu, and J. Jia, “Deep convolutional neural network for image deconvolution,” in Advances in Neural Information Processing Systems (NIPS), 2014.
    Google ScholarLocate open access versionFindings
  • S. Nah, T. H. Kim, and K. M. Lee, “Deep multi-scale convolutional neural network for dynamic scene deblurring,” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • S. Su, M. Delbracio, J. Wang, G. Sapiro, W. Heidrich, and O. Wang, “Deep video deblurring,” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    Google ScholarLocate open access versionFindings
  • T. M. Nimisha, A. Kumar Singh, and A. N. Rajagopalan, “Blurinvariant deep learning for blind-deblurring,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • T. H. Kim, K. M. Lee, B. Scholkopf, and M. Hirsch, “Online video deblurring via dynamic temporal blending network,” in The IEEE International Conference on Computer Vision (ICCV), 2017.
    Google ScholarLocate open access versionFindings
  • T. H. Kim, S. Nah, and K. M. Lee, “Dynamic video deblurring using a locally adaptive linear blur model,” IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017.
    Google ScholarLocate open access versionFindings
  • J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
    Google ScholarLocate open access versionFindings
  • I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets,” in Advances in Neural Information Processing Systems (NIPS), 2014.
    Google ScholarLocate open access versionFindings
  • A. Radford, L. Metz, and S. Chintala, “Unsupervised representation learning with deep convolutional generative adversarial networks,” International Conference on Learning Representations (ICLR), 2016.
    Google ScholarLocate open access versionFindings
  • A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, “Vision meets robotics: The kitti dataset,” The International Journal of Robotics Research, vol. 32, no. 11, pp. 1231–1237, 2013.
    Google ScholarLocate open access versionFindings
  • A. Sellent, C. Rother, and S. Roth, “Stereo video deblurring,” in European Conference on Computer Vision (ECCV), 2016. Kaihao Zhang is currently pursuing the Ph.D. degree with the College of Engineering and Computer Science, The Australian National University, Canberra, ACT, Australia. Prior to that, he received the M.Eng. degree in computer application technology from the University of Electronic Science and Technology of China, Chengdu, China, in 2016. He worked at the Center for Research on Intelligent Perception and Computing, National Lab of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, China for two years and the Tencent AI Laboratory, Shenzhen, China for one year. His research interests focus on video analysis and facial recognition with deep learning.
    Google ScholarLocate open access versionFindings
  • Wenhan Luo is currently working as a senior researcher in the Tencent AI Lab, China. His research interests include several topics in computer vision and machine learning, such as motion analysis (especially object tracking), image/video quality restoration, reinforcement learning. Before joining Tencent, he received the Ph.D. degree from Imperial College London, UK, 2016, M.E. degree from Institute of Automation, Chinese Academy of Sciences, China, 2012 and B.E. degree from Huazhong University of Science and Technology, China, 2009.
    Google ScholarLocate open access versionFindings
  • Yiran Zhong received the M.Eng in information and electronics engineering in 2014 with the first class honor from The Australian National University, Canberra, Australia. After two years of research assistant, he becomes a PhD student in the College of Engineering and Computer Science, The Australian National University, Canberra, Australia and Data61, CSIRO, Canberra, Australia. He won the ICIP Best Student Paper Award in 2014. His current research interests include geometric computer vision, machine learning and deep learning.
    Google ScholarLocate open access versionFindings
  • Lin Ma (M13) received the B.E. and M.E. degrees in computer science from the Harbin Institute of Technology, Harbin, China, in 2006 and 2008, respectively, and the Ph.D. degree from the Department of Electronic Engineering, The Chinese University of Hong Kong, in 2013. He was a Researcher with the Huawei NoahArk Laboratory, Hong Kong, from 2013 to 2016. He is currently a Principal Researcher with the Tencent AI Laboratory, Shenzhen, China. His current research interests lie in the areas of computer vision, multimodal deep learning, specifically for image and language, image/video understanding, and quality assessment. Dr. Ma received the Best Paper Award from the Pacific-Rim Conference on Multimedia in 2008. He was a recipient of the Microsoft Research Asia Fellowship in 2011. He was a finalist in HKIS Young Scientist Award in engineering science in 2012.
    Google ScholarLocate open access versionFindings
  • Hongdong Li is currently a Reader with the Computer Vision Group of ANU (Australian National University). He is also a Chief Investigator for the Australia ARC Centre of Excellence for Robotic Vision (ACRV). His research interests include 3D vision reconstruction, structure from motion, multiview geometry, as well as applications of optimization methods in computer vision. Prior to 2010, he was with NICTA Canberra Labs working on the Australia Bionic Eyes project. He is an Associate Editor for IEEE T-PAMI, and served as Area Chair in recent year ICCV, ECCV and CVPR. He was a Program Chair for ACRA 2015 - Australia Conference on Robotics and Automation, and a Program Co-Chair for ACCV 2018 - Asian Conference on Computer Vision. He won a number of prestigious best paper awards in computer vision and pattern recognition, and was the receipt for the CVPR Best Paper Award in 2012 and the Marr Prize Honorable Mention in 2017.
    Google ScholarLocate open access versionFindings
您的评分 :
0

 

标签
评论
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn
小科