Jingwen Guo

Jinwen Guo (郭静文) is a master student in the School of Eltronic and Computer Engineering, Peking University (PKU). She is supervised by Prof. Hong Liu at Open Lab on Human Robot Interaction (HRI). Her research interests include computer vision, action recognition, self-supervised learning, and federated learning.

Email  /  ZhiHu  /  Google Scholar  /  Github

I'm interested in computer vision, action recognition, self-supervised learning, and federated learning.

We pioneer a novel Federated Skeleton-based Action Recognition (FSAR) paradigm, which enables the construction of a globally generalized model without accessing local sensitive data.

We propose a Camera-style Separation and Uncertainty Estimation (CSUE) framework for Unsupervised domain adaptation person re-identification tasks.

A navigation method based on Semantic Spatial Relationships (SSR) is proposed and is shown to have more reliable performance when dealing with target-driven visual navigation.

We proposed a Fire-Station Location Planning model based on Genetic Algorithm (GAFLP) to effectively solve the problems of high construction cost and large space resource waste of urban fire stations.

Driving strategies are trained from reinforcement learning simulators using generative adversarial networks as well as imitation learning to improve the per-right and robustness of autonomous driving.

Implementation of a system for the development of mapping retrieval and visualisation systems.

Repetitive work for designers is replaced by AI: automatically generates the corresponding boundary wireframes for different types of images, reducing the designer's workload in creating line drawings.