An empirical evaluation for defining a mid-air gesture dictionary for web-based interaction

Overview

• This paper evaluates the use of mid-air gestures for web-based interactions, such as searching for a university classroom.
• The researchers worked with 56 HCI students who designed 99 different mid-air gestures for 16 web-based actions.
• They then tested these gestures with 248 users to identify the most intuitive and recurring gestures, as well as potential issues.
• The researchers propose a mid-air gesture dictionary that contains the most suitable gestures for each web-based action.

Plain English Explanation

The paper explores the use of mid-air gestures, where users interact with a computer system without touching a physical device, in the context of web-based applications. The researchers worked with a group of 56 HCI (human-computer interaction) students, who were tasked with identifying the main actions required for a web-based university classroom search service and then proposing mid-air gestures to carry out those actions.

The students came up with a total of 99 different mid-air gestures for 16 different web-based actions. Then, the researchers had 248 external users test and validate these gesture proposals. By analyzing the results, the researchers were able to identify the most intuitive and commonly used mid-air gestures, as well as potential issues that may arise when using these gestures for web-based interactions.

Based on their findings, the researchers developed a mid-air gesture dictionary, which they believe contains the most suitable gestures for each web-based action identified in the study. The key insight from this research is that people tend to use gestures that are similar to those employed in touch-based and mouse-based interfaces, even though these may not always be the most effective for mid-air interactions due to the difference in the user's distance from the device.

Technical Explanation

The study was conducted in three main stages. First, the 56 HCI student designers worked in 16 separate groups to identify the key actions required for a web-based university classroom search service. They then proposed a set of mid-air gestures to carry out these actions, resulting in a total of 99 different mid-air gestures.

In the second stage, the researchers involved 248 external users to validate the gesture proposals made by the student designers. These users were asked to evaluate the intuitive nature and suitability of the mid-air gestures for the specified web-based actions.

Finally, the researchers analyzed the results of the user validation process to identify the most recurring or intuitive mid-air gestures, as well as potential issues with the proposed gestures. Based on this analysis, they defined a mid-air gesture dictionary that contains the most suitable gestures for each of the identified web-based actions.

The key finding from this research is that users tend to rely on gestures that are similar to those used in touch-based and mouse-based interfaces, even though these may not always be the most effective for mid-air interactions. This is due to the difference in the distance between the user and the device in each interaction context.

Critical Analysis

The paper provides a valuable empirical evaluation of mid-air gestures for web-based interactions, which is an important area of research as we see an increasing shift towards touchless interfaces. However, there are a few limitations and areas for further exploration that could be considered.

First, the study was conducted with a relatively small sample size of 56 student designers and 248 users. While this provides a good starting point, expanding the study to a larger and more diverse group of participants could yield additional insights and help validate the findings.

Additionally, the study focused on a specific web-based application (university classroom search) and may not be directly applicable to other web-based contexts. Further research could explore the use of mid-air gestures in a broader range of web-based applications to assess the generalizability of the findings.

The researchers also mention the potential criticalities associated with the proposed mid-air gestures, but do not delve deeply into the specific issues or potential solutions. Exploring these in more detail could provide valuable guidance for designers and developers working on mid-air gesture-based web interfaces.

Overall, this paper makes a valuable contribution to the understanding of mid-air gestures in web-based interactions, but there are opportunities to build upon this work and address some of the limitations identified.

Conclusion

This research paper presents an empirical evaluation of mid-air gestures for web-based interactions, focusing on a university classroom search service. The study involved HCI students designing mid-air gestures for various web-based actions, which were then validated by external users.

The key findings suggest that users tend to rely on gestures similar to those used in touch-based and mouse-based interfaces, even though these may not always be the most effective for mid-air interactions due to the difference in the user's distance from the device. The researchers propose a mid-air gesture dictionary that contains the most suitable gestures for each identified web-based action.

This research provides valuable insights for designers and developers working on mid-air gesture-based web interfaces, and highlights the need for further exploration of this emerging interaction paradigm. By understanding the intuitiveness and potential issues associated with mid-air gestures, the field can continue to evolve and deliver more seamless and natural web-based experiences for users.