新闻

  • [2021.07.15更新] 有关实验室22级硕士研究生招生,详情请点击 研究生招生说明
  • [2021.07.15更新] 实验室招募本科实习生(长期),详情请点击 本科生招生说明
  • [2021.02.13更新] 实验室拟招聘1-2名博士后研究人员,详情请点击 博士后招聘说明
  • 2021.07.19:程先航视频插帧论文被IEEE TPAMI接收。
  • 2021.06.11:李楠楠黑盒攻击论文被IEEE TIP接收。
  • 2021.05.27:2021届硕士生毕业,将加入阿里、腾讯、Intel等单位工作或前往EPFL等高校深造。
  • 2021.05.25:吴伟、丁晓颖、刘子政、柳费洋博士毕业,将分别加入阿里、高校、腾讯、百度。
  • 2021.05.11:2021.05.11:实验室陈境远,丁冠辰,杨雨辰等人团队在CVPR2021 AICity比赛Track 4以较小分差排名百度,字节跳动之后位列第三。
  • 2021.03.19:柳费洋视频编码论文被IEEE TIP接收。
  • 2021.01.11:实验室与诺基亚贝尔实验室等单位共同制定的虚拟现实国际标准ITU-T P.919正式发布。
  • 2020.12.04:陈震中教授受邀为《遥感学报》第六届编委。
  • 2020.12.01:孙万捷、李一鸣博士毕业,将分别加入武汉大学、腾讯。
  • 2020.11.03:刘子政视频编码论文被IEEE TIP接收。
  • 2020.10.14:张考视频显著性论文被IEEE TIP接收。
  • 2020.09.09:陈震中教授受邀为SCI一区/CCF A类期刊IEEE TIP编委(AE)(影响因子9.34)。
  • 2020.08.31:刘锐论文被IEEE TGRS接收。
  • 2020.07.30:欧阳君论文被IEEE Trans. CSVT接收。
  • 2020.06.17:朱耀晨论文被IEEE Trans. Affective Computing接收。

更多新闻 ...

最新成果

  • 20207.22
    In this work, we propose a novel black-box attack approach that can directly minimize the induced distortion by learning the noise distribution of the adversarial example, assuming only loss-oracle access to the black-box network. Our attack results in low distortion as validated on ImageNet. Read more...
  • 20206.15
    In this paper, we propose an enhanced deformable separable convolution (EDSC) based network to obtain information from non-local neighborhood than is capable to produce multiple in-between frames. Experimental results show that our method performs favorably against the state-of-the-art methods. Read more...
  • 20205.27
    In this work, a novel 6DoF mesh saliency database is developed. We also propose a 6DoF mesh saliency detection algorithm together with an evaluation metric for 6DoF experiments accordingly. Moreover, some state-of-the-art saliency detection methods have been extended as benchmarks. Read more...
  • 20203.28
    In this paper, we propose a multimodal variational encoder-decoder (MMVED) framework for micro-video popularity prediction tasks. MMVED learns a stochastic Gaussian embedding of a micro-video that is informative to its popularity level while preserves the inherent uncertainties simultaneously. Read more...
  • 20203.24
    In this paper, we propose a refined Adversarial Inverse Reinforcement Learning (rAIRL) method to handle the reward ambiguity problem by disentangling reward for each word in a sentence, as well as achieve stable adversarial training by refining the loss function to shift the generator towards Nash equilibrium. Read more...