KR101855105B1 - Focus point analysis system for human eye on head-mounted display - Google Patents

Abstract

A method for analyzing a focal position of an eye includes the steps of: (a) separating focus data of a plurality of eyes into N analysis zones; (b) calculating a dense center point for each of the N analysis zones, for each analysis zone in which the focus data exists; and (c) calculating a dense center point between the analysis zones by using the dense center point of the corresponding analysis zone and a dense center point of the analysis zone adjacent to the corresponding analysis zone, for the analysis zone representing density of a preset level or more among the dense center points for each analysis zone of the step (b). According to the present invention, a point with high interest of a user can be detected by analyzing the focal position of the user in a head mounted display.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system and method for analyzing a focus position of an eye in a head mount display,

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a system and method for analyzing a focus position of an eye in a head mount display, and more particularly, To a system and method for analyzing the focal position of an eye in a display.

Japanese Patent Application Laid-Open No. 10-2016-0022927 (eye tracking system for head mount display) and the like discloses a system capable of detecting a user's line of sight position on a head mount display.

However, even if the eye line, which is the focus position of the user's eyes, may be fixed at one position, it may be moved or fixed at one position. Instead of gazing at one point, May appear in the form of a high density of focus positions.

The focal point position information of the user's eyes can be utilized variously by being used for analyzing the behavior pattern of the user or the like. For example, the PPL effect may be increased when the user actually looks at the PPL picture in the image. In this case, the advertisement cost for the PPL may be additionally requested.

An object of the present invention is to solve the technical problem as described above, and it is an object of the present invention to provide a head mount display capable of detecting a point of high interest by analyzing a focus position of an eye in a head mount display. And a method for analyzing the focal position of the focal point.

According to a preferred embodiment of the present invention, an eye focal position analyzing method includes: (a) separating focus data of a plurality of eyes into N analysis regions; (b) calculating a dense center point for each of the N analysis regions, for each analysis region, in which the focus data exists; And (c) for the analysis area having density higher than a certain level among the density center points of the analysis areas in the step (b), using the density center point of the analysis area and the density center point of the analysis area adjacent to the analysis area, And calculating a dense center point.

Specifically, the dense center point of each analysis zone in the step (b) is a position at which a distance between focus data in one analysis zone is minimized.

In addition, in the step (c), it is preferable to calculate the dense center point between the analysis zones using the dense center information of the dense central point and the dense information of the adjacent deny zone as the weight information.

In addition, the method for analyzing the focal point position of eyes according to the present invention may further include grouping successive image information into image information units of a first time before the step (a). Preferably, the steps (a) to (c) are performed for each group grouped in the grouping step, wherein at least some of the image information in one group is grouped before or after the group As shown in FIG.

In addition, the plurality of eye focus data in the step (a) is characterized in that a straight line extending in the eye direction of the camera of the head mount display worn by the user collides with the projection object on which the image is projected .

According to the system and method for analyzing the focal position of the eye in the head mount display of the present invention, it is possible to detect a point of high interest by the user by analyzing the focal position of the eye in the head mount display.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a configuration diagram of an eye focus position analysis system in a head mount display according to a preferred embodiment of the present invention; FIG.
2 is an explanatory diagram of a focus position acquisition and transmission method according to the present invention;
FIGS. 3A and 3B are explanatory diagrams of a focal position detection method by a tracer according to the present invention, respectively.
4 is a flowchart of a method of analyzing the focal position of eyes by the analyzer according to an embodiment of the present invention.
5 is an explanatory diagram of a step S10 of grouping.
6 is an explanatory diagram of a step S20 of separating a plurality of focus data into N analysis regions.
7 is an explanatory diagram of a step (S30) of calculating a dense center point for each analysis area.
8 is an explanatory diagram of a step (S40) of calculating a dense center point between analysis zones.
FIG. 9 is an explanatory diagram of a result of the focus position analysis method of the present invention. FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a system and method for analyzing an eye focal position in a head mount display according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

It should be understood that the following embodiments of the present invention are only for embodying the present invention and do not limit or limit the scope of the present invention. It will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

FIG. 1 is a block diagram of a system 100 for analyzing the focal position of an eye in a head mounted display according to a preferred embodiment of the present invention.

As can be seen from FIG. 1, the eye focus position analysis system 100 preferably comprises a head mount display 110 and a server 120.

The head mount display 110 includes a renderer 111 and a tracer 112.

The renderer 111 detects the focal position of the eye and plays a role of drawing video, metadata and CG (Computer Graphic) objects on the head mount display 110.

In addition, the tracer 112 detects the focal position of the user's eye using the head-mounted display 110 and collects the detected focal position data. The collected focus position data is transmitted from the head mount display 110 to the server 120.

2 is an explanatory diagram of a focus position acquisition and transmission method of the present invention.

As can be seen from FIG. 2, in the present invention, the detected focal position is stored in a temporary storage (memory) together with a tag (detection time information) that can determine the time of the detection time, Seconds, 5 seconds, etc.) to the server 120 at once. This may vary depending on the load of the server 120 and the performance of the client and may be transmitted to the server 120 upon detection of the focus position.

3A and 3B are explanatory diagrams of a focal position detection method by the tracer 112 of the present invention, respectively.

Specifically, FIG. 3A is an explanatory view of the focus position detection in the two-dimensional image projection, and FIG. 3B is an explanatory view of the focus position detection in the 180/360 degree image projection.

As can be seen from FIGS. 3A and 3B, the focus position is determined by collision with a projection object (Virtual Screen) on which an image that can be expressed by a straight line extending in the camera eye direction and a cube, hemisphere, Dimensional coordinates of the point.

That is, a point at which a straight line extending in the eye direction of the camera of the head mount display 110, which the user wears, collides with the projection object on which the image is projected is detected as the focus position. Specifically, a straight line extending in the visual line direction of the camera becomes the center line of the user's visual field.

The server 120 of the present invention may include a data storage unit 121 and an analysis unit 122.

The data received from the head mount display 110 is stored in the data storage unit 121 and then analyzed by the analysis unit 122. In addition, the analysis result data that has been analyzed by the analysis unit 122 may be stored in the data storage unit 121 or may be stored in a separate storage.

The focal position data stored in the data storage unit 121 can be divided into two types. The moving coordinate in focus and the fixed coordinate, which is the fixed focus coordinate, are finally fixed. However, fixed focus coordinates and moving focus coordinates can be distinguished by density. The fixed focal point coordinates are generally concentrated within a certain area, and when moving, they have a specific directionality. Since the data of interest at present is a fixed focal point coordinate, it should be determined whether it is dense or not. In the present invention, density is determined using a cluster algorithm to determine such density. A clustering algorithm is an algorithm that maximizes the homogeneity of unit elements within a group by maximizing similarity among data and maximizes the difference between groups. The clustering algorithms can use various methods according to data characteristics, structure, and analysis purpose. Among them, it is considered to be most suitable to use the density based analysis algorithm.

4 is a flowchart illustrating a method of analyzing the focal position of an eye by the analyzer 122 according to an embodiment of the present invention. The analysis unit 122 may be implemented by at least a portion of at least one processor included in the server 120. [ That is, the eye focal position analysis method of the present invention can be implemented by at least a part of one processor.

As can be seen from FIG. 4, the method for analyzing the focal position of an eye according to the preferred embodiment of the present invention includes the steps of grouping continuous image information into image information units of a first time (S10) (S20) of dividing the focus data of a plurality of eyes into N analysis regions, a step (S30) of calculating a dense center point for each of the analysis regions, and an analysis region (S40) calculating a dense center point between the analysis zones using the dense center point of the analysis zone and the dense center point of the analysis zone adjacent to the analyzed zone.

Fig. 5 is an explanatory diagram of the grouping step S10.

Since the focus extracted by the head mount display 110 is focused on the image, the frame moves naturally as the frame of the image is changed. Therefore, it is not meaningful to grasp the focal position with respect to the whole focal data, and it is preferable to analyze the focal position in groups of as short a time unit as possible. The time unit may vary depending on the characteristics of the image, which means that it may be a different value depending on the characteristic rather than the fixed specific period. In addition, since the shift of the focus is continuous and related to the previous position, the analysis accuracy is increased by adding a certain portion of overlapping sections. For example, in the second-second analysis, the analysis is performed in units of 0 to 2 seconds including the previous 0 to 1 second interval.

That is, in the grouping step (S10) of the present invention, at least some image information in one group is used in a group before or after the group.

6 is an explanatory diagram of a step S20 of separating a plurality of focus data into N analysis regions. In Fig. 6, sixteen analysis zones using a rectangle are illustrated by way of example.

In step S20, it is preferable that a figure such as a circle or a rectangle is implemented as one analysis zone. However, since the present invention is based on image information, it is more preferable to separate the analysis area by using a rectangle proportional to the image frame rather than the circular shape.

That is, when density analysis is performed on the entire data, more accurate density measurement can be performed, but the speed may be lowered. Therefore, it is expected that the analysis speed can be improved by dividing the analysis area according to the size of the image and calculating the density of the focus data in each analysis area in parallel. The size of the analysis area may vary depending on the resolution of the image.

7 is an explanatory diagram of a step (S30) of calculating a dense center point for each analysis area.

The dense center point of each analysis zone in step S30 is characterized by being calculated by a position that minimizes the distance between the focus data in one analysis zone. In addition, in step S30, it is preferable to perform the density analysis on the remaining area except the analysis area where no data exists after the analysis area is created.

That is, in step S30, it is preferable to calculate the dense center point for each analysis area only for the analysis area where the focus data exists among the N analysis areas. In addition, the density can be calculated as an average of the distance of each focus data with respect to the center point.

In the case of density analysis in the analysis area, the calculation speed is fast, but as can be seen in the analysis area 8 and the analysis area 12 in FIG. 7, it is difficult to find the exact position when dividing into both areas, and the problem is divided into two center points.

Accordingly, in step S40, the center point of the correct cluster is regenerated by applying the K-Means Algorithm based on each center point generated in the analysis area. Also, only when the density is higher than the reference value, it is analyzed only to increase the calculation speed.

FIG. 8 is an explanatory diagram of a step (S40) of calculating a dense center point between analysis zones.

For analysis zone 6, it is necessary to calculate the dense center point for analysis zones 1, 2, 3, 5, 7, 9, 10, 11, However, in the case of analysis zones 1, 2, 3, 5, 7, 9, 10, and 11, the density is less than the reference value and is excluded from the calculation of the density center point between analysis zones. .

In addition, in the case of the analysis zone 8, it is necessary to calculate the density center point for the analysis zones 3, 4, 7, 11, 12 including the analysis zone 8. In the case of the analytical zones 3, 4, 7 and 11, since the density is less than the reference value, it is excluded from the calculation of the dense center point between the analytical zones. Therefore, by using the dense center point of the analytical zone 8 and the dense central point of the analytical zone 12, .

Similarly, in the case of the analysis zone 12, it is necessary to calculate the dense center point for the analysis zones 7, 8, 11, 15, 16, including the analysis zone 12. In the case of the analysis zones 7, 11, 15 and 16, since the density is less than the standard value, it is excluded from the calculation of the density center point between the analysis zones. Therefore, by using the density center point of the analysis zone 8 and the density center point of the analysis zone 12, .

Specifically, the dense center point between analysis zones is calculated by a position that minimizes the distance between dense center points between analysis zones.

Although no weight is applied in FIG. 8, depending on the case, in step S40, the density center point between the analysis zones and the density information of the adjacent analysis zones are used as the weight information to calculate the density center point between the analysis zones You can do it.

FIG. 9 shows an explanatory diagram of the results of the focus position analysis method of the present invention.

When 50 pieces of focus data are displayed on a two-dimensional plane, they have various distributions as shown in FIG. Intuitively, it can be seen that the density of the fixed focus is high and the density of the moving focus is low. According to the eye focal position analysis system 100 and method of the head mount display 110 of the present invention, a clustering algorithm for finding a fixed focus of interest and removing a moving focal point out of the focus data, It can be seen that the fixed focus can be extracted by using

As described above, according to the eye focal point position analyzing system 100 and method thereof in the head-mounted display 110 of the present invention, by analyzing the focal point position of the eye in the head-mounted display 110, It can be seen that a point having a high degree of interest can be detected.

100: Focus position analysis system
110: Head mount display
120: Server
111: The renderer
112: Tracer
121: Data storage unit
122:

Claims (7)

(a) separating focus data of a plurality of eyes into N analysis regions;
(b) calculating a dense center point for each of the N analysis regions, for each analysis region, in which the focus data exists; And
(c) For the analysis area having a density higher than a certain level among the density center points of the analysis area in the step (b), the density center point of the analysis area and the density center point of the analysis area adjacent to the analysis area are used, Calculating a center point,
Wherein the dense center point of each analysis region in the step (b) is a position at which a distance between focus data in one analysis region is minimized. delete delete The method according to claim 1,
The step (c)
Wherein the dense center point between the analysis zones is calculated using the dense center information of the dense analysis zones and the dense information of the adjacent analysis zones as the weight information. The method according to claim 1,
The method of claim 1,
Before the step (a)
And grouping successive image information into image information units of a first time period. 6. The method of claim 5,
Performing the steps (a) to (c) for each group grouped in the grouping step,
Wherein at least some of the image information in one group is used overlappingly in a group before or after the corresponding group. The method according to claim 1,
The plurality of eye focus data in the step (a)
Wherein a straight line extending in a direction of a line of sight of a camera of a head mounted display worn by a user collides with a projection object on which an image is projected.

Images