Abstract

This work focuses on improving remote experimentation and data collection by addressing its current underdeveloped state, which often results in the exclusion of participants due to technical issues. To overcome such issues, implementing quality controls has the potential to improve remote experimentation. To this effect, we have implemented quality controls in an open source project - the experimental hub. The platform enables researchers to conduct remote experiments across various fields such as interpersonal synchrony, social adaptive user interfaces, WebAR filters, and more. Through customization, experimenters can tailor the hub to their specifications and incorporate additional features that enhance their remote experiments. In the experimental hub, a customized filter for detecting eyeglasses has been created to address problems with certain computer vision algorithms that do not perform optimally when glasses are being used. Next, dynamic filters were implemented to facilitate experimenters' ability to adjust the filters to their needs. This implementation simplified the developer workflow to allow future developers to easily create filters and integrate them into the frontend through an automated generation process. Finally, the project's wiki was updated to provide guidance for new developers. This thesis offers recommendations and guidelines to increase the project's status, broaden participation among diverse developers, and enhance the development ecosystem.

Results/Implementation/Project Description

Conclusion

[ PDF (optional) ] 

[ Slides Kickoff/Final (optional)]