I am currently working as a senior researcher at Fujitsu Research in Japan. I graduated from Nanjing University (China) with a bachelor’s degree and obtained my master’s and doctor’s degrees from Nara Institute of Science and Technology (Japan).

My research interests include

Multi-camera motion capturing
Pose estimation
Action recognition

Please feel free to email me at hongheyangfan@gmail.com.

🔥 News

2023.10: 🎉 Our paper was selected for the IROS Best Paper Award Finalist on Cognitive Robotics.
2023.10: 🎉 Our Gymnastic Judging Support System has been successfully applied in 10 apparatuses at the 52nd FIG Artistic Gymnastics World Championships. I was responsible for developing the multi-camera gymnast detection and tracking modules for our system. Parts of the related techniques have been published in my CVM2023 and WACV2023 papers, with more details to come in my journals currently under review.

📝 Publications

More papers and reports are shown on my Google Scholar.

Multi-camera Gymnasst Tracking Module In Fujitsu Gymnastic Judging Support System

Fujitsu Gymnastic Judging Support System

Our Gymnastic Judging Support System has been successfully applied in 10 apparatuses at the 52nd FIG Artistic Gymnastics World Championships. I was responsible for developing the multi-camera gymnast detection and tracking modules for our system. Multiple related patents have been applied (not online yet):
- “Cross-view Bounding Box Interpolation” (2021)
- “Semi-supervised Multi-frame Multi-view Athlete Detection” (2022)
- “Collaborative Muti-frame Multi-view Triangulation” (2022)
- “Hybrid Multi-camera Athlete Tracking” (2022)
More details to come in my journals currently under review.

🎥🤸🤼 Action Recognition

IROS 2023 Is Weakly-supervised Action Segmentation Ready For Human-Robot Interaction? No, Let’s Improve It With Action-union Learning | (Best Paper Award Finalists on Cognitive Robotics)|Fan Yang, Shigeyuki Odashima, Shoichi Masui, Shan Jiang.
arXiv 2022Actor-identified Spatiotemporal Action Detection | Fan Yang, Norimichi Ukita, Sakti Sakriani, Nakamura Satoshi code
ACM MM ASIA 2021 Make Skeleton-based Action Recognition Model Smaller, Faster and Better |Fan Yang, Sakti Sakriani, Wu Yang, Nakamura Satoshi | code
IEEE ACCESS 2021 A framework for knowing who is doing what in aerial surveillance videos |Fan Yang, Wu Yang, Sakti Sakriani, Nakamura Satoshi

🎥👣 Multi-object Tracking

CVM Journal 2023 A Unified Multi-view Multi-person Tracking Framework, |Fan Yang, Shigeyuki Odashima, Sosuke Yamao, Hiroaki Fujimoto, Shoichi Masui and Shan Jiang
WACV 2023 Hard to Track Objects with Irregular Motions and Similar Appearances? Make It Easier by Buffering the Matching Space, |Fan Yang, Shigeyuki Odashima, Shoichi Masui, Shan Jiang
Image and Vision Computing 2022 Tackling multiple object tracking with complicated motions—re-designing the integration of motion and appearance |Fan Yang, Xin Chang, Yang Wu, Sakriani Sakti, Satoshi Nakamura
Image and Vision Computing 2021 ReMOT: A model-agnostic refinement for multiple object tracking |Fan Yang, Xin Chang, Yang Wu, Sakriani Sakti, Satoshi Nakamura
ICASSP 2020Using panoramic videos for multi-person localization and tracking in a 3D panoramic coordinate |Fan Yang, Feiran Li, Yang Wu, Sakriani Sakti, Satoshi Nakamura| code

🎥Pose Estimation

IEEE ACCESS 2019 A soft proposal segmentation network (SPS-Net) for hand segmentation on depth videos |Fan Yang, Wu Yang, Sakti Sakriani, Nakamura Satoshi
CVPR 2018Depth-based 3d hand pose estimation: From current achievements to future goals | Shanxin Yuan, …, Fan Yang, …, Tae-Kyun Kim

Others

Machine Intelligence Research 2022Efficient Visual Recognition with Deep Neural Networks: A Survey on Recent Advances and New Directions | Yang Wu, Dingheng Wang, Xiaotong Lu, Fan Yang, Guoqi Li, Weisheng Dong, Jianbo Shi
Computers \& Electrical Engineering 2021Representation-guided generative adversarial network for unpaired photo-to-caricature translation | Ziqiang Zheng, Hongzhi Liu, Fan Yang, et al.
IEEE Transactions on Multimedia 2020Instance-level Heterogeneous Domain Adaptation for Limited-labeled Sketch-to-Photo Retrieval | Fan Yang, Wu Yang, Wang Zheng, Li Xiang, Sakti Sakriani, Nakamura Satoshi | code

IEEE EMBC 2018Application of SsVGMM to medical data-classification with novelty detection| Fan Yang, Soriano Jaymar, Kubo Takatomi, Ikeda Kazushi | code

ICONIP 2017A hierarchical mixture density network| Fan Yang, Soriano Jaymar, Kubo Takatomi, Ikeda Kazushi

🎖 Honors and Awards

Conference Awards

2023 IROS Best Paper Award Finalist on Cognitive Robotics.
Challenges Awards
2022 ECCV Multiple Object Tracking and Segmentation in Complex Environments Workshop Challenges: 🥈2nd-place award on multiple people tracking in group dance task.
2022 CVPR SoccerNet & ActivityNet Workshops Challenges: 🥈2nd-place award on multi-camera multi-person tracking task.
2021 ICCV Multi-camera Multi-person Tracking Workshop Challenges: 🏅4th-place award on multi-camera multi-person tracking task. Click the image to see the report on YouTube.
2020 CVPR MOTS Workshop Challenges: 🥇1st-place award on multi-object tracking and segmentation task.
2020 CVPR Workshop on Autonomous Driving (WAD) Challenges: 🥇1st-place award on BDD100K multi-object tracking task.
2017 ICCV Workshop on 3D Hand Pose Estimation Challenges: 🥇1st-place award on the object interaction task and 🥈2nd-place award on the tracking task. Click the image to see the report on YouTube.