Resume
Download a PDF copy of my resume on the right (might not be up-to-date).
Basics
| Name | Gueter Josmy Faure |
| Label | Computer Vision Researcher |
| josmyfaure@gmail.com | |
| Phone | (+886) 909402049 |
| Url | https://joslefaure.github.io |
| Summary | Computer Vision Researcher/Engineer who likes to force machines to understand videos. I am working towards a Ph.D. in Computer Science at the National Taiwan University. |
Work
-
2022.08 - Present Senior Engineer - Computer Vision
HTC
Teaching tiny computers to accurately estimate human poses. This includes hand poses (VR headsets), body poses (vision-based trackers) and facial expressions (avatars).
- Virtual Reality
- Body/hand/face Tracking
- Computer Vision
-
2022.07 - 2022.02 Computer Vision Engineer Intern
Movella
Help develop the company's first markerless tracker (Android/iOS app) for monitoring physically challenged patients and assist in their rehabilitation.
- Android Development
- Computer Vision
- Body Pose Tracking
-
2019.08 - 2022.01 AI Engineer Intern
艾鍗科技
Develop computer vision algorithms for resource-constrained hardware (Micro:Bit, SparkFun Edge, Raspberry Pi...). Design Computer vision courses tailored for kids.
- Computer Vision
- Embedded Systems
- AI
Education
-
2023.09 - Present Taipei, Taiwan
-
2020.09 - 2022.08 Hsinchu, Taiwan
-
2016.09 - 2020.06 Hsinchu, Taiwan
Awards
- 2022.12.02
Master's Thesis Excellence Award
Taiwanese Association of Artificial Intelligence (TAAI)
This Award goes to the authors of the best Master's Theses on Artificial Intelligence in Taiwan.
- 2022.10.23
ECCV 2022 DeeperAction Challenge
European Conference on Computer Vision (ECCV)
I led the team which placed second. We designed a multimodal action detection framework to accurately detect multiple persons performing multiple overlapping actions.
- 2020.06.01
Taiwan AI CUP Honorable Mention
Ministry of Education, Taiwan (台灣教育部)
Our group submitted one of the best solutions for that year's AI CUP.
Certificates
| Tensorflow Developer Certificate | ||
| Google, Tensorflow | 2021-04 |
| Deep Learning Specialization | ||
| Deeplearning.ai | 2019-08 |
Publications
-
2024.10.01 HERMES: temporal-coHERent long-forM understanding with Episodes and Semantics
Arxiv
Extended our ECCVW paper, BREASE with more datasets and more ablations
-
2024.08.24 Bridging Episodes and Semantics: A Novel Framework for Long-Form Video Understanding
ECCVW
Took inspiration from cognitive processes to develop an episodic memory module and a semantic parser to efficiently understand long videos.
-
2023.09.18 Interaction-Aware Prompting for Zero-Shot Spatio-Temporal Action Detection
ICCV
We propose the first Zero-Shot Spatio-Temporal Action Detection model.
-
2023.02.10 Holistic Interaction Transformer Network for Action Detection
WACV
Develop a model that uses skeleton information to improve Action Detection.
-
2021.11.9 Confidence-Aware Anomaly Detection in Human Actions
ACPR
We propose to use the pose feedback to improve anomaly detection.
Skills
| Artificial Intelligence | |
| Deep Learning | |
| Computer Vision | |
| Multimodal Language Model | |
| Video Understanding | |
| Deep Learning Frameworks (PyTorch, TensorFlow) |
Languages
| French | |
| Fluent |
| English | |
| Fluent |
| Chinese | |
| Advanced |
References
| Professor Winston Hsu | |
| My PhD advisor. The king of high-level vision. His abhorence for incremental research is my greatest inspiration. |
| Professor Shang-Hong Lai | |
| My Master's advisor. His insistence that we publish makes it really hard not to. |