Skip to content

Commit a328195

Browse files
committed
fiddle with pictures
1 parent 77fa614 commit a328195

File tree

4 files changed

+2
-2
lines changed

4 files changed

+2
-2
lines changed

assets/images/mas_dataset.png

389 KB
Loading

assets/images/old_mas_dataset.png

1.45 MB
Loading

assets/images/stress_detector.png

-1.83 MB
Binary file not shown.

projects.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,7 @@ I curated a dataset of 353 short clips of people in monologue scenarios experien
1111
<br>
1212

1313
## Multimodal Stress Detector
14-
![Multimodal Stress Detector](assets/images/stress_detector.png)
14+
![Multimodal Stress Detector](assets/images/icmi2024.png)
1515
<br><br>
1616
I designed an audio-visual stress detector that uses late fusion to interpret expressions of stress across multiple modalities. This approach allows the system to combine the strength of multiple modalities while maintaining robust performance even when some are unavailable. This system achieved an F1 score of 0.85 on the MAS dataset and is currently being investigated for use in detecting and correcting implicit bias
1717

@@ -29,4 +29,4 @@ I helped write the control architecture of Ommie, a therapy robot being deployed
2929
## Real-Time Course Feedback
3030
![Course Feedback](assets/images/course_feedback.png)
3131
<br><br>
32-
I led a team of undergrads to design a system that allowed students to give feedback on a class in real-time using a Kinect body-tracking system installed outside the lecture hall. Out of 36 teams, our project won the class choice award for CPSC 484: Human-Computer Interaction.
32+
I led a team of undergrads to design a system that allowed students to give feedback on a class in real-time using a Kinect body-tracking system installed outside the lecture hall. Professors were also able to view aggregate statistics about their class, helping them to improve their teaching lecture-to-lecture. Out of 36 teams, our project won the class choice award for CPSC 484: Human-Computer Interaction.

0 commit comments

Comments
 (0)