adhd indicators modelling based on dynamic time warping from rgb data: a feasibility study
DESCRIPTION
ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study. Antonio Hernández-Vela, Miguel Reyes, Laura Igual, Josep Moya, Verónica Violant , and Sergio Escalera. ADHD: Attention deficit hyperactivity disorder. Inattention. Hyperactivity. Impulsivity. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/1.jpg)
ADHD indicators modelling based on Dynamic Time
Warping from RGB data: A feasibility study
Antonio Hernández-Vela, Miguel Reyes, Laura Igual, Josep Moya,
Verónica Violant, and Sergio Escalera
![Page 2: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/2.jpg)
2
ADHD: Attention deficit hyperactivity disorder
Inattention
Hyperactivity
Impulsivity
![Page 3: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/3.jpg)
3
Outline
1. Introduction
2. Methodology
3. Results
4. Conclusion
![Page 4: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/4.jpg)
4
Introduction
• Video-based behavior analysis for ADHD diagnosis in children between 8-11 years.
• Automatic detection of ADHD visual indicators
![Page 5: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/5.jpg)
5
Introduction
• Behavior analysis Human pose information along time
HeadBodyHands
time Gestures
Inattention
Hyperactivity
Impulsivity
1. Data acquisition2. Feature extraction: Human Pose3. Gesture detection
![Page 6: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/6.jpg)
6
Outline
1. Introduction
2. Methodology
1. Data acquisition
2. Feature extraction
3. Gesture detection
3. Results
4. Conclusion
![Page 7: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/7.jpg)
7
Data aqcuisition
Microsoft’s Kinect
RGB + Depth
• Invariant to color, texture and lighting conditions• Human pose directly obtained
![Page 8: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/8.jpg)
8
Feature extraction: Human Pose
RGB + Depth Body skeleton
• 42-dimensional vector: 14 joints × 3 spatial dimensions
![Page 9: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/9.jpg)
9
Gesture detection
• Dynamic Time Warping (DTW)
![Page 10: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/10.jpg)
10
Threshold computing
• Leave-one-out similarity measure between different samples and gestures
G1G11
G12
…
G13
G2G21
G22
…
G23
… GnGn1
Gn2
Gn3
G11
Different gestures
Differentsamples
![Page 11: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/11.jpg)
11
Outline
1. Introduction
2. Methodology
3. Results
4. Conclusion
![Page 12: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/12.jpg)
12
Results
![Page 13: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/13.jpg)
13
Results
![Page 14: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/14.jpg)
14
Outline
1. Introduction
2. Methodology
3. Results
4. Conclusion
![Page 15: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/15.jpg)
15
Outline
1. Introduction
2. Methodology
3. Results
4. Conclusion
![Page 16: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/16.jpg)
16
Conclusion
• Methodology for gesture segmentation and recognition at the same time.
• First results indicate the objectives are feasible.
Future work:• Automatic callibration• Feature weighting (body joints)
![Page 17: ADHD indicators modelling based on Dynamic Time Warping from RGB data: A feasibility study](https://reader035.vdocuments.mx/reader035/viewer/2022062408/56813e68550346895da878b9/html5/thumbnails/17.jpg)
ADHD indicators modelling based on Dynamic Time
Warping from RGB data: A feasibility study
Antonio Hernández-Vela, Miguel Reyes, Laura Igual, Josep Moya,
Verónica Violant, and Sergio Escalera
Thank You!
Questions?