您当前的位置:
情感计算组
组 长: 曾加贝 副研究员;山世光 研究员
Email: jiabei.zeng [at] vipl dot ict dot ac dot cn; sgshan [at] ict dot ac dot cn
课题组简介

  情感计算组致力于从计算机视觉的角度,研究机器学习等方法感知并深刻理解人类的意图、情感和精神状态,主要包括情绪类型的识别、人脸动作单元检测、情感正负性和强度估计、心理状态和精神状况评估等。情感计算组在国内外重要期刊和会议发表论文若干,并多次获得国际学术竞赛冠亚军。

研究内容


  

  该研究组的研究内容包括面向视觉情感感知的学习方法、视觉情感计算基本任务、应用三个方面:


  (1)面向视觉情感计算的学习方法
  重点针对情感计算数据和任务的特点,研究非理想标注条件下的机器学习算法,如自/半监督学习、弱监督学习、协同监督学习等基础理论与方法。

  (2)基于视觉的情感计算基本任务

     面部表情相关

           离散情绪分类

         

           面部动作单元(facial action unit, AU)检测

       

           情感维度估计

                   

          o 零样本多标签表情识别、基于视觉的心理状态估计(如专注、无聊、紧张、焦虑等)

         视线方向相关

          o 三维视线方向估计

          o 二维屏幕视点估计

          o 视线跟踪与视线模式判断

      (3)基于视觉的情感计算应用

         金融信贷评估中的应用
         精神状况评估的客观化(儿童孤独症、抑郁症等)
         驾驶员状态评估

    部分论文

    刊物论文

    • Mengyi Liu, Shaoxin Li, Shiguang Shan, Xilin Chen, \"AU-inspired Deep Networks for Facial Expression Feature Learning,\" Neurocomputing, vol. 159, pp. 126-136, 2015.
    • Mengyi Liu, Ruiping Wang, Shaoxin Li, Zhiwu Huang, Shiguang Shan, Xilin Chen, \"Video Modeling and Learning on Riemannian Manifold for Emotion Recognition in the Wild,\" Journal on Multimodal User Interfaces, vol. 10, no. 2, pp. 113-124, 2016.
    • Mengyi Liu, Ruiping Wang, Shiguang Shan, Xilin Chen, \"Learning Prototypes and Similes on Grassmann Manifold for Spontaneous Expression Recognition,\" Computer Vision and Image Understanding, vol. 147, pp. 95-101, 2016.
    • Mengyi Liu, Shiguang Shan, Ruiping Wang, Xilin Chen, \"Learning Expressionlets via Universal Manifold Model for Dynamic Facial Expression Recognition,\" IEEE Transactions on Image Processing, vol. 25, no. 12, pp. 5920-5932, 2016.
    • Yong Li, Jiabei Zeng and Shiguang Shan. Learning Representations for Facial Actions from Unlabeled Videos. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2020. (Accepted)

    会议论文

    • Mengyi Liu, Xin Liu, Yan Li, Xilin Chen, Alexander Hauptmann, Shiguang Shan, \"Exploiting Feature Hierarchies With Convolutional Neural Networks for Cultural Event Recognition,\" IEEE International Conference on Computer Vision(ICCV2015) Workshops, pp. 32-37, 2015.
    • Jiabei Zeng, Shiguang Shan, Xilin Chen, \"Facial Expression Recognition with Inconsistently Annotated Datasets,\" To appear in Proc. European Conference on Computer Vision (ECCV2018), pp. 1-16, Munich, Germany, 2018.
    • Zijia Lu, Jiabei Zeng, Shiguang Shan, Xilin Chen, \"Zero-Shot Facial Expression Recognition with Multi-Label Label Propagation,\" Asian Conference on Computer Vision 2018(ACCV2018), 2-6 Dec. 2018, Perth Western Australia
    • Yong Li, Jiabei Zeng, Shiguang Shan, Xilin Chen, \"Self-supervised Representation Learning from Videos for Facial Action Unit Detection\", IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp.10924-10933, Long Beach, California, USA, June 16-20, 2019.
    • Yunjia Sun, Jiabei Zeng, Shiguang Shan, Xilin Chen. Cross-Encoder for Unsupervised Gaze Representation Learning. IEEE/CVF International Conference on Computer Vision (ICCV), pp. 3702-3711, Montreal, Canada, Oct. 11-17, 2021.
    • Xin Cai, Jiabei Zeng and Shiguang Shan. Landmark-aware Self-supervised Eye Semantic Segmentation. Unknown Aware Feature Learning for Face Forgery Detection. International Conference on Automatic Face and Gesture Recognition (FG), 2021. (Accepted)
    • Xuran Sun, Jiabei Zeng and Shiguang Shan. Emotion-aware Contrastive Learning for Facial Action Unit Detection. International Conference on Automatic Face and Gesture Recognition (FG), 2021. (Accepted)