Towards Better Throwing: A Comparison of Performance and Preferences Across Point of Release Mechanics in Virtual Reality
Letter Of Changes: Dear reviewers,
Thank you for your thorough comments and for validating the timeliness of our submission. The majority of fixes were clarifications, however, we also acknowledged certain limitations and adjusted the implications of our results accordingly. Specifically, we expanded on reasons for using Kinect and details about latency and double-exponential filter (Sec. 3.2.5) while pointing out its underperformance as part of the 'External Sensor' condition [R WPij, R t12p, R q9Fv], and highlighted that future work should use other sensors since if accuracy is good enough, users may prefer this condition due to not having any trackers on their body (Sec. 6).
### General changes
* CH and CP relied on analog button input to acquire/release the virtual objects [R WPij, R q9Fv], so to ensure that this did not confound the study, we had the training phase of the study as described in Sec. 3.9, allowing participants to get comfortable with all conditions before data collection. This ensured that the actual evaluation of throwing between different conditions was the same. For clarity, we added this detail in Sec. 3.2.
* We revised classifications in Figure 1 and Sec. 3.1 [R t12p, R q9Fv] by including an extra Manual PoR detection through on-HMD hand tracking. We initially based the taxonomy on PoR detection approaches executed in the literature, briefly mentioning this one in Sec 3.2, Sec 3.7, and Sec. 6, however, as [R t12p] pointed out, this input configuration can become mainstream once hand tracking FOV is no longer an issue.
* To describe input standardization between HTC Vive and Kinect [R WPij, R q9Fv], we added the following to Sec. 3.2.5: "Given that coordinate system origins of Vive HMD and Kinect were different, the system computed the offset between the user's head and hand joints in Kinect's coordinates, then added this offset to the Vive HMD position each frame."
* Meaning of thresholds [R WPij]: threshold always refers to the distance at which the tracked joint must be away from the defined origin for PoR to be detected. During calibration, these thresholds were adjusted following the procedure in Sec. 3.3 and Table 1 (based on participant's arm span). We added this clarification to Section 3.9.
* The differences between CT and VT threshold values [R WPij, R q9Fv] came from different devices tracking different joints -- those that worked best for each device (palm for Kinect, wrist for VT, handheld controller for CT). While visually the ball always appeared on the hand, detection of threshold crossing depended on the tracking device. We added this detail in Sec 3.3.
### Figure 1 changes [R t12p, R q9Fv]
* Added ‘On-HMD hand tracking with gesture recognition’ under Manual PoR and adjusted related text in Sec. 3.1. [R t12p]
* Improved clarity in examples and caption (to clarify how IMU and hand tracking are under multiple PoR detection approaches) [R t12p]
### Writing, formatting, grammar, references
* Consolidated the two accuracy figures into a single Figure 4 [R t12p, R q9Fv]
* Defined PoR in the introduction [R t12p, R q9Fv]
* Changed 2 conditions -> 2 independent variables (Sec. 3.6) [R t12p, R q9Fv]
* Changed navigation -> manipulation (Section 1) [R t12p, R q9Fv]
* The references that are made to websites are now updated to provide the URL information as [R t12p] requested, these references are from forums and online discussions created by VR players through recent years, this issue was also stated in “Performance and Experience of Throwing in Virtual Reality” by Zindulka et al. (2020) cited in Sec. 1 [R q9Fv].
* APA citation style [R q9Fv]: when using LaTeX, we found that using \documentclass[acmtog,anonymous,review]{acmart} was the way to get closest to APA format, however updated instructions on GI website (https://graphicsinterface.org/conference/2024/formatting-instructions/) ask to use \documentclass[sigconf, screen, anonymous]{acmart}. We were not sure what to follow, could you please clarify the appropriate way to make citations look right?
Thank you,
Authors
Keywords: Virtual Reality, Input Devices, Body Tracking, Imperical User Study
Abstract: An underexplored interaction metaphor in virtual reality (VR) is throwing, with a considerable challenge in achieving accurate and natural results. We conducted an empirical investigation of participants' performance in a VR throwing task, measuring their accuracy and preferences across Point of Release (PoR) mechanics (manual and automatic) with various input device categories (hand-held, on-body, external) and throwable object types. Participants were tasked with throwing a baseball, a bowling ball, and a football toward targets using 5 input configurations (2 manual and 3 automatic PoR).
Results from 30 participants indicate that the overall highest accuracy was achieved with an automatic PoR configuration (on-body tracker). The post-study and VR survey results indicate that the majority of participants preferred a manual PoR configuration (hand-held VR controller-derived) for the throwing direction, throwing speed, and as being the closest to real-life throwing. Our findings are useful for VR researchers and developers who want to implement throwing as a technique in their applications.
Submission Number: 22
Loading