Limits...
Modelling human visual navigation using multi-view scene reconstruction.

Pickup LC, Fitzgibbon AW, Glennerster A - Biol Cybern (2013)

Bottom Line: Participants viewed a simple environment from one location, were transported (virtually) to another part of the scene and were asked to navigate back.We also measured error distributions when participants manipulated the location of a landmark to match the preceding interval, providing a direct test of the landmark-location stage of the navigation models.Models such as this, which start with scenes and end with a probabilistic prediction of behaviour, are likely to be increasingly useful for understanding 3D vision.

View Article: PubMed Central - PubMed

Affiliation: School of Psychology and Clinical Language Sciences, University of Reading, Reading, RG6 6AL, UK. l.c.pickup@reading.ac.uk

ABSTRACT
It is often assumed that humans generate a 3D reconstruction of the environment, either in egocentric or world-based coordinates, but the steps involved are unknown. Here, we propose two reconstruction-based models, evaluated using data from two tasks in immersive virtual reality. We model the observer's prediction of landmark location based on standard photogrammetric methods and then combine location predictions to compute likelihood maps of navigation behaviour. In one model, each scene point is treated independently in the reconstruction; in the other, the pertinent variable is the spatial relationship between pairs of points. Participants viewed a simple environment from one location, were transported (virtually) to another part of the scene and were asked to navigate back. Error distributions varied substantially with changes in scene layout; we compared these directly with the likelihood maps to quantify the success of the models. We also measured error distributions when participants manipulated the location of a landmark to match the preceding interval, providing a direct test of the landmark-location stage of the navigation models. Models such as this, which start with scenes and end with a probabilistic prediction of behaviour, are likely to be increasingly useful for understanding 3D vision.

Show MeSH

Related in: MedlinePlus

Four examples of the pole-position models, corresponding to the four sets of pole and goal positions in Fig. 2. The shapes represent the hypothesized uncertainties over pole locations. Note that the covariances vary according to the distance from the pole to the viewing strip (heavy magenta line). In each model, the pole positions are recorded in egocentric coordinates, here represented by the thin - and -axis, so the coordinate frame is independent of the coordinate frame of the room for each condition
© Copyright Policy - OpenAccess
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3755223&req=5

Fig3: Four examples of the pole-position models, corresponding to the four sets of pole and goal positions in Fig. 2. The shapes represent the hypothesized uncertainties over pole locations. Note that the covariances vary according to the distance from the pole to the viewing strip (heavy magenta line). In each model, the pole positions are recorded in egocentric coordinates, here represented by the thin - and -axis, so the coordinate frame is independent of the coordinate frame of the room for each condition

Mentions: The model builds up a reconstruction of the 3-pole scene in an egocentric coordinate frame by assuming there is a set of cameras all pointing at the central (green) pole and the cameras lie in a strip that extends a distance along the -axis (where here  cm), as shown in Fig. 3. and are free parameters in the model. This mirrors the configuration of the “start zone” in interval one which allowed for 80 cm of free motion left and right along an axis perpendicular to the direction in which the green (central) pole lay, while allowing minimal motion in depth (up to  cm). Participants were asked to step side to side within the start zone. We used the above parameters in the reconstruction model. Since all the information about the 3D room can be captured in its 2D plan view, we consider 1D images of this 2D space, instead of 2D images of the whole 3D virtual environment. The “image noise” on any one of these hypothetical 1D measurements is taken to be Gaussian with a standard deviation of , i.i.d. for each measurement.Fig. 3


Modelling human visual navigation using multi-view scene reconstruction.

Pickup LC, Fitzgibbon AW, Glennerster A - Biol Cybern (2013)

Four examples of the pole-position models, corresponding to the four sets of pole and goal positions in Fig. 2. The shapes represent the hypothesized uncertainties over pole locations. Note that the covariances vary according to the distance from the pole to the viewing strip (heavy magenta line). In each model, the pole positions are recorded in egocentric coordinates, here represented by the thin - and -axis, so the coordinate frame is independent of the coordinate frame of the room for each condition
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3755223&req=5

Fig3: Four examples of the pole-position models, corresponding to the four sets of pole and goal positions in Fig. 2. The shapes represent the hypothesized uncertainties over pole locations. Note that the covariances vary according to the distance from the pole to the viewing strip (heavy magenta line). In each model, the pole positions are recorded in egocentric coordinates, here represented by the thin - and -axis, so the coordinate frame is independent of the coordinate frame of the room for each condition
Mentions: The model builds up a reconstruction of the 3-pole scene in an egocentric coordinate frame by assuming there is a set of cameras all pointing at the central (green) pole and the cameras lie in a strip that extends a distance along the -axis (where here  cm), as shown in Fig. 3. and are free parameters in the model. This mirrors the configuration of the “start zone” in interval one which allowed for 80 cm of free motion left and right along an axis perpendicular to the direction in which the green (central) pole lay, while allowing minimal motion in depth (up to  cm). Participants were asked to step side to side within the start zone. We used the above parameters in the reconstruction model. Since all the information about the 3D room can be captured in its 2D plan view, we consider 1D images of this 2D space, instead of 2D images of the whole 3D virtual environment. The “image noise” on any one of these hypothetical 1D measurements is taken to be Gaussian with a standard deviation of , i.i.d. for each measurement.Fig. 3

Bottom Line: Participants viewed a simple environment from one location, were transported (virtually) to another part of the scene and were asked to navigate back.We also measured error distributions when participants manipulated the location of a landmark to match the preceding interval, providing a direct test of the landmark-location stage of the navigation models.Models such as this, which start with scenes and end with a probabilistic prediction of behaviour, are likely to be increasingly useful for understanding 3D vision.

View Article: PubMed Central - PubMed

Affiliation: School of Psychology and Clinical Language Sciences, University of Reading, Reading, RG6 6AL, UK. l.c.pickup@reading.ac.uk

ABSTRACT
It is often assumed that humans generate a 3D reconstruction of the environment, either in egocentric or world-based coordinates, but the steps involved are unknown. Here, we propose two reconstruction-based models, evaluated using data from two tasks in immersive virtual reality. We model the observer's prediction of landmark location based on standard photogrammetric methods and then combine location predictions to compute likelihood maps of navigation behaviour. In one model, each scene point is treated independently in the reconstruction; in the other, the pertinent variable is the spatial relationship between pairs of points. Participants viewed a simple environment from one location, were transported (virtually) to another part of the scene and were asked to navigate back. Error distributions varied substantially with changes in scene layout; we compared these directly with the likelihood maps to quantify the success of the models. We also measured error distributions when participants manipulated the location of a landmark to match the preceding interval, providing a direct test of the landmark-location stage of the navigation models. Models such as this, which start with scenes and end with a probabilistic prediction of behaviour, are likely to be increasingly useful for understanding 3D vision.

Show MeSH
Related in: MedlinePlus