Application-Driven AI Paradigm for Human Action Recognition

no code implementations • 30 Sep 2022 • Zezhou Chen, Yajie Cui, Kaikai Zhao, Zhaoxiang Liu, Shiguo Lian

Human action recognition in computer vision has been widely studied in recent years.

Action Classification Action Recognition +2

A Realistic Face-to-Face Conversation System based on Deep Neural Networks

no code implementations • 21 Aug 2019 • Zezhou Chen, Zhaoxiang Liu, Huan Hu, Jinqiang Bai, Shiguo Lian, Fuyuan Shi, Kai Wang

Based on the models' output, the synthesizer uses the Pixel2Pixel model to generate realistic facial images.

Generative Adversarial Network

A Neural Virtual Anchor Synthesizer based on Seq2Seq and GAN Models

no code implementations • 20 Aug 2019 • Zipeng Wang, Zhaoxiang Liu, Zezhou Chen, Huan Hu, Shiguo Lian

This paper presents a novel framework to generate realistic face video of an anchor, who is reading certain news.

Word Embeddings

Facial Pose Estimation by Deep Learning from Label Distributions

no code implementations • 30 Apr 2019 • Zhaoxiang Liu, Zezhou Chen, Jinqiang Bai, Shaohua Li, Shiguo Lian

Facial pose estimation has gained a lot of attentions in many practical applications, such as human-robot interaction, gaze estimation and driver monitoring.

Gaze Estimation Pose Estimation