Modeling Pointing for 3D Target Selection in VR

Dalsgaard, Tor-Salve; Knibbe, Jarrod; Bergström, Joanna

doi:10.1145/3489849.3489853

Cited by 14 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The modeling approach for human pointing establishes the use of various motion and encoded features as effective inputs for inferring target directions. [8].…”

Section: A Modeling Human Pointing At Target Fixationmentioning

confidence: 99%

“…Previous works on pointing pose instructed participants to mark point completion time either by manual controller inputs or encoded user motion [1], [7], [8]. This instruction does not fit our application of collecting natural pointing movements from start to finish.…”

Section: Data Collection Proceduresmentioning

confidence: 99%

“…Previous studies on inferring human pointing targets [6]- [8] use tracked position and orientation of body parts, such as the head, torso, and hands. Among the most intuitive of such features are those rooted in the principles of ray-casting, where a virtual ray is defined by an object orientation or between two reference points, originating from arms, hands, or eyes [9]- [12].…”

Section: Introductionmentioning

confidence: 99%

“…Among the most intuitive of such features are those rooted in the principles of ray-casting, where a virtual ray is defined by an object orientation or between two reference points, originating from arms, hands, or eyes [9]- [12]. Ray-casting vector, while not as accurate as eye-trackers, provide useful information for modeling human pointing at fixation pose, and for classification and inference of pointing targets [7], [8], [13]. However, its use is not widely explored for early-prediction of object interactions, as is the case for kinematic joint information.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Continuous Prediction of Pointing Targets With Motion and Eye-Tracking in Virtual Reality

Chung,

Lee

2024

IEEE Access

View full text Add to dashboard Cite

We present a study on continuously predicting the direction to a pointing target in virtual environments using motion and eye-tracker data throughout the pointing process. We first collect time series data for user motion and eye-tracker in a cursorless, single-target pointing task. Results from analyzing fixation points from different sensors and observing velocity profiles over the course of pointing provide insights into optimally configuring features for predicting the target angles. Following this analysis, we train a recurrent neural network that feeds on sliding window inputs for continuously operating target direction prediction from start to finish. The input window contains historical data from past to current frames, capturing temporal changes in the feature data. By feeding on this input, our model can predict the direction of the target at any given time during pointing. Our findings demonstrate that incorporating eye-tracker data into the prediction model boosts the maximum achievable accuracy by 2.5 times when compared to baselines without eye-tracker data inputs. The results suggest that using features from both the eye-tracker and joint motion contributes to higher prediction performance, as well as faster stabilization of output values at the starting phase of pointing.

show abstract

“…The modeling approach for human pointing establishes the use of various motion and encoded features as effective inputs for inferring target directions. [8].…”

Section: A Modeling Human Pointing At Target Fixationmentioning

confidence: 99%

Section: Data Collection Proceduresmentioning

confidence: 99%