Multimodal utterance-level affect analysis using visual, audio and text features
Deng, Didan, et al. "Multimodal utterance-level affect analysis using visual, audio and text features." arXiv preprint arXiv:1805.00625 (2018).
CV
Education
- B.S. in Electrical Engineering, Fudan University, 2017
- Ph.D in Electronic Engineering and Computer Science, HKUST, 2022 (expected)
Skills
- Python, Matlab
- PyTorch, Tensorflow
Publications
Unsupervised Outlier Detection in Appearance-Based Gaze Estimation
Chen, Zhaokang, et al. "Unsupervised Outlier Detection in Appearance-Based Gaze Estimation." Proceedings of the IEEE International Conference on Computer Vision Workshops. 2019.
MIMAMO Net: Integrating Micro-and Macro-motion for Video Emotion Recognition
Deng, Didan, et al. "MIMAMO Net: Integrating Micro-and Macro-motion for Video Emotion Recognition." arXiv preprint arXiv:1911.09784 (2019).
Multitask Emotion Recognition with Incomplete Labels
Deng, Didan, Zhaokang Chen, and Bertram E. Shi. "Multitask Emotion Recognition with Incomplete Labels." 2020 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020)(FG).
Towards Better Uncertainty: Iterative Training of Efficient Networks for Multitask Emotion Recognition
Deng, Didan, Wu Liang, and Bertram E. Shi. "Towards Better Uncertainty: Iterative Training of Efficient Networks for Multitask Emotion Recognition" 2021 International Conference on Computer Vision Workshop.
Ensembling With a Fixed Parameter Budget: When Does It Help and Why?
Deng, Didan and Bertram E. Shi. "Ensembling With a Fixed Parameter Budget: When Does It Help and Why?" 2021 13th Asian Conference on Machine Learning.