About

Yunzhong Hou (侯云钟) is a research fellow at the Australian National University, collaborating closely with Prof. Tom Gedeon and Prof. Liang Zheng. He completed his PhD at the same university in 2023, under the esteemed guidance of Prof. Liang Zheng, Prof. Stephen Gould, and Prof. Hongdong Li. Yunzhong’s academic journey began at Tsinghua University, where he earned a bachelor’s degree in electronic engineering in 2018.

Specializing in computer vision and deep learning, his current research interest lies in AI-based drone videography, camera control and optimizations for embodied agents, and multi-view detection and tracking. Yunzhong is actively involved in the academic community, contributing as a conference reviewer for prominent conferences such as CVPR, ECCV, ICCV, NeurIPS, ACM MM, and IJCAI. He also serves as a journal reviewer for respected publications like TPAMI, TIP, TMM, and TCSVT, and an area chair for ICASSP and ACM MM.

For more details, please find his CV here.

News

• 2023.12 Our grant proposal “Privacy-Percerving Perception for Robotics” is awarded by the HMI seed grant for 25,000 AUD! Big thank you to Dylan, Rahul, and Mike!

• 2023.12 Check out our latest research on camera layout optimization “Optimizing Camera Configurations for Multi-View Pedestrian Detection”. arxiv

• 2023.11 Our paper “View-Coherent Correlation Consistency for Semi-Supervised Semantic Segmentation” is accepted by Pattern recognition. paper.

• 2023.06 I was named outstanding reviewer for CVPR 2023!

• 2023.04 Joined as a research fellow at ANU, working with Prof. Tom Gedeon and Dr. Liang Zheng. Excited!

• 2023.03 Check out our latest research “Learning to Select Camera Views: Efficient Multiview Understanding at Few Glances” on arXiv. paper, code

• 2022.07-2022.10 Internship at Amazon Web Services as a research scientist on vision-language tasks. Hello Bay Area!

• 2021.12 Our paper “Adaptive Affinity for Associations in Multi-Target Multi-Camera Tracking“ is accepted by IEEE TIP. paper, code

• 2021.07 Our paper “Ranking Models in Unlabeled New Environments“ is accepted by ICCV 2021. paper, code

• 2021.07 Our paper “Multiview Detection with Shadow Transformer (and View-Coherent Data Augmentation)“ is accepted by ACM MM 2021. paper, code

• 2021.03 Our paper “Visualizing Adapted Knowledge in Domain Transfer“ is accepted by CVPR 2021. paper, code, 知乎-UDA可视化, 知乎-无需风格图像的风格迁移.

• 2020.07 Our paper “Multiview Detection with Feature Perspective Transformation“ is accepted by ECCV 2020. paper, code, 知乎, MultiviewX dataset download.

• 2020.03 Our paper “Learning to Structure an Image with Few Colors“ is accepted by CVPR 2020. paper, code, 知乎.

• 2019.11 A new paper “Locality aware appearance metric for multi-target multi-camera tracking” is released on arXiv. paper, code, 知乎.

• 2019.06 Won 5th place out of 22 participants in multi-target multi-camera tracking in CVPR 2019 AI-City Challenge. paper, code.

• 2019.06 Won 3rd place out of 84 participants in vehicle re-identification in CVPR 2019 AI-City Challenge. paper, code.

• 2019.03 Our paper “Improving Device-Edge Cooperative Inference of Deep Learning via 2-Step Pruning” is accepted by Infocom workshop on IECOO 2019. paper, code.

Contact

Please contact me via e-mail.