- Time: 13:30 - 14:00
- Attendees: Bob Zhang (Supervisor), Huang Yanzhen, Mai Jiajun
- Enriched Dataset using a smaller sampling step from the videos to perform a more formal model performance evaluation with test set that's separated from the train & evaluation set.
- Exposed interface for face cropping.
Poster preparation:
- Documented poster content and discussed with supervisor about the further edition.
- Made some polishing on poster structure.
- Translation of poster of Chinese version. Interview:
- Be talking about our project and campus life then. Rehearsal:
- Test device on Friday.
- Displayed latest demo by online video and camera using OBS.
- Enhanced performance for stream-pushing by directly altering DOM instead of using react states; as well as giving away image quality in python backend.
- Discussed about how to write interim report.
- Check hands key points' confidence before doing phone detection.
- Update str above the rectangles in demo.
- Brainstorming on further dataset.
- Face detection.