CN103472923A - Method for selecting scene objects with three-dimensional virtual gesture - Google Patents

Method for selecting scene objects with three-dimensional virtual gesture Download PDF

Info

Publication number
CN103472923A
CN103472923A CN2013104453769A CN201310445376A CN103472923A CN 103472923 A CN103472923 A CN 103472923A CN 2013104453769 A CN2013104453769 A CN 2013104453769A CN 201310445376 A CN201310445376 A CN 201310445376A CN 103472923 A CN103472923 A CN 103472923A
Authority
CN
China
Prior art keywords
candidate
gesture
desktop
dimensional
vector
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2013104453769A
Other languages
Chinese (zh)
Other versions
CN103472923B (en
Inventor
冯志全
张廷芳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Jinan
Original Assignee
University of Jinan
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Jinan filed Critical University of Jinan
Priority to CN201310445376.9A priority Critical patent/CN103472923B/en
Publication of CN103472923A publication Critical patent/CN103472923A/en
Application granted granted Critical
Publication of CN103472923B publication Critical patent/CN103472923B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a method for selecting scene objects with a three-dimensional virtual gesture, and the method comprises the following steps that (1) a three-dimensional gesture model library is established, and is characterized by also comprising the following steps that (2) a desktop equation of a placed object is determined; (3) the initial time T is equal to 0; (4) the movement track of the three-dimensional gesture is predicted, and a straight line l is predicted through the position information HT and HT+1 of the time T and T+1 of the three-dimensional gesture; (5) an object set C1 to be selected is initially selected; (6) an object set C2 to be selected is finally determined; (7) the object to be selected is displayed, and candidate objects in the set C2 are displayed at highlight; (8) a collision test is taken to the object to be selected; the collision test is taken to each candidate object in the object set C2 in sequence; if the collision test succeeds, the test is over; otherwise, T is equal to T+1, T is the time, and the step (4) is carried out. According to the method for selecting the scene objects with the three-dimensional virtual gesture, the objects in a scene are selected accurately with the three-dimensional gesture.

Description

A kind of three-dimensional gesture is selected the method for object scene
Technical field
The present invention relates to three-dimensional gesture field, specifically, relate to a kind of method that three-dimensional gesture is selected object scene.
Background technology
Selecting object by indication is the simplest a kind of action in reality, as long as the object that a given choice direction can be chosen by intersecting the test judgement.In immersive environment, often with the position of viewpoint and virtual hand, form choice direction, and, under desktop environment, only need the position of mouse point to get final product.This technology can be traced back to MIT " put-that-there " interface exploitation in 1980 the earliest, wherein with voice, confirms to select to handle.Afterwards, a large amount of selection technology are developed, and the difference of these technology comes from this two design variables: the mode that choice direction forms and the scope of selection (determining to select to have how many objects possibilities selected) at every turn.
Select basically not need user's physical action owing to giving directions, therefore only just select it can obtain good user performance.Yet, because this indication technology of dimensional limit is not suitable for the object positioning action, the user only can locate on the radian around viewpoint, object can not be moved along choice direction.
Ray-Casting the most simply gives directions the selection technology.Current choice direction is determined by the position h that is attached to vector P on virtual hand and virtual hand: p (a)=h+ap, and 0<a<+∞.This technology can be selected the object of distant place arbitrarily in theory, but, along with the increase of distance and dwindling of object size, the difficulty of choosing is also increasing.Due to the impact of hardware shake, the error rate of choosing also improves gradually simultaneously.
" searchlight " selecting technology (Spotlight) replaces original ray with one from the cone of selecting the vector starting point to start, as long as there have object to drop in this cone to be selected, be similar to object and be illuminated the same.This technology is selected especially favourable for small items, but also there will be a plurality of objects to drop on the situation in cone simultaneously, can increase maloperation on the contrary.
Aperture is the improvement technology to Spotlfight, and it allows user's interactively when by cone, choosing object to control the scope of projection.The projecting direction of cone determines by the tracker of head and virtual hand, and the user can interactively adjusts the distance of virtual hand and head and adjusts the cone size, with this, eliminates the ambiguousness of selection.
The thought of Image-Plane comes from a ultimate principle in graphics: any object that we see is all in the perspective projection of looking on the pyramid back plane, therefore only needs a point on given projection plane just can be selected.This technology is the same with several technology in front can not change the distance of alternative from the operator, Fishing-Ree1 has made up this shortcoming, by extra increase input equipment, as increased button or sliding bar on tracker, the user can move selecteed object along the selection vector by interactively, but, because the six-freedom degree by object separates control, so still can reduce user performance.
Summary of the invention
The technical scheme that the present invention will solve is to provide a kind of method that three-dimensional gesture is selected object scene, realizes that three-dimension gesture accurately selects the object in scene.
The present invention adopts following technical scheme to realize goal of the invention:
A kind of three-dimensional gesture is selected the method for object scene, comprises the steps:
(1) set up the three-dimension gesture model storehouse;
It is characterized in that, also comprise the steps:
(2) determine the desktop equation of the object of putting;
(3) initialization moment T=0;
(4) movement locus of prediction three-dimension gesture, utilize three-dimension gesture T and T+1 positional information H constantly tand H t+1, prediction straight line l:
The equation of straight line l is: x = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) x T + ( t - T ) / ( ( T + 1 ) - T ) x T + 1 y = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) y T + ( t - T ) / ( ( T + 1 ) - T ) y T + 1 z = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) z T + ( t - T ) / ( ( T + 1 ) - T ) z T + 1 ,
T=0,1,2 ..., n, the wherein frame number of n for predicting, wherein H t=(x t, y t, z t), H t+1=(x t+1, y t+1, z t+1);
(5) tentatively selected object collection C1 to be selected;
(6) finally determine object collection C2 to be selected;
(7) show object to be selected, the candidate's object in set C2 is carried out to high light display and show;
(8) object to be selected is carried out to collision detection, successively each candidate's object of object collection C2 is carried out to collision detection, if the collision detection success finishes, otherwise, T=T+1, T is constantly, goes to step (4).
As the further restriction to the technical program, described step (2) comprises the steps:
(2.1) three-dimensional coordinate of desktop lower-left angle point A in three-dimensional scenic, and it is known to intersect at the vector on Liang Tiao limit, the desktop lower left corner, i.e. two crossing vectors on desktop
Figure BDA0000385789370000032
known;
(2.2) by an A, vector
Figure BDA0000385789370000033
obtain the plane equation of desktop: Z=f (x, y, z).
As the further restriction to the technical program, described step (5) comprising:
(5.1) setting threshold value to be chosen is R, wherein threshold value
Figure BDA0000385789370000034
m ifor the coordinate of object i, the number of the last object of N desktop;
(5.2) obtain an intersection point o by straight line l and desktop Z;
(5.3) calculate successively the distance of each object to intersection point o:
Figure BDA0000385789370000035
r ithe object of<R is candidate's object, joins the set collection C of candidate's object 1in.
As the further restriction to the technical program, described step (6) comprising:
(6.1) calculated candidate object C successively 1the cosine value of the angle between each object of set and the normal vector of hand; Cos<n, H t+1-M i, the normal vector that n is hand;
(6.2) judgement cos<n, H t+1-M iwhether be greater than 0, if be greater than 0, this object is candidate's object, joins candidate's object collection C 2in.This step is exactly that object is got rid of at candidate's object of hand back.
Compared with prior art, advantage of the present invention and good effect are: the reasonable selection technology of 1, mentioning in background will be by some equipment, and as increased button or sliding bar on tracker, and the present invention need not.2, the selection technology of mentioning in background is only selected according to the position of object, and the present invention is according to the status information of gesture, and the selection of carrying out object of object relative position real-time update in hand and scene.3, the selection technology of mentioning in background is substantially all remote selection, chooses the probability of object also to be subject to the distance of object, size impact, as more difficult as object smaller in scene choosing.The present invention closely selects, and when hand is closer apart from object, just chooses, and so just not there will be the situation of the difficult choosing of wisp, selects degree of accuracy high.
The accompanying drawing explanation
Fig. 1 is processing flow chart of the present invention.
Embodiment
Below in conjunction with accompanying drawing and preferred embodiment, the present invention is further described in detail.
Referring to Fig. 1, the three-dimensional gesture refers to employing based on computer vision technique, utilization tracking or the reconstructing method computing machine is that generate, consistent with user's gesture three-dimension gesture model, and this is prior art, sets up the step in three-dimension gesture model storehouse in this omission.Utilize the three-dimensional gesture to select the method for object scene as follows:
Step1. determine the desktop equation of the object of putting, the three-dimensional coordinate of desktop lower-left angle point A in three-dimensional scenic, and it is known to intersect at the vector on Liang Tiao limit, the desktop lower left corner, i.e. two crossing vectors on desktop
Figure BDA0000385789370000051
Figure BDA0000385789370000052
known; By an A, vector
Figure BDA0000385789370000053
obtain the plane equation of desktop: Z=f (x, y, z).
Step2. initialization T=0 constantly;
Step3. dope the curve movement of three-dimensional hand, i.e. the movement locus of three-dimensional hand.Utilize three-dimensional hand T and T+1 positional information H constantly tand H t+1, prediction straight line l:
The equation of straight line l is: x = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) x T + ( t - T ) / ( ( T + 1 ) - T ) x T + 1 y = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) y T + ( t - T ) / ( ( T + 1 ) - T ) y T + 1 z = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) z T + ( t - T ) / ( ( T + 1 ) - T ) z T + 1 ,
T=0,1,2 ..., n, the wherein frame number of n for predicting, wherein H t=(x t, y t, z t), H t+1=(x t+1, y t+1, z t+1).
Step4. tentatively select object collection to be selected.Setting threshold value to be chosen is R, by straight line l and desktop Z, obtains an intersection point o, calculates successively the distance of each object to intersection point o;
Figure BDA0000385789370000055
r ithe object of<R is candidate's object, joins the set C of candidate's object 1in.Wherein the set threshold value of this paper is
Figure BDA0000385789370000056
Figure BDA0000385789370000057
m ifor the coordinate of object i, the number of the last object of N desktop.
Step5. finally determine object collection to be selected, calculated candidate object C successively 1the cosine value of the angle between each object of set and the normal vector of hand; Cos<n, H t+1-M i.Cos<n, H t+1-M i0 object is candidate's object, joins the set C of candidate's object 2in.This step is exactly that object is got rid of at candidate's object of hand back.Cos<n wherein, H t+1-M i0, now candidate's object is at the front of hand, cos<n, H t+1-M i><0, now candidate's object is in the back of hand.The normal vector that n is hand.
Step6. show object to be selected.To gather C 2in candidate's object carry out high light display and show.
Step7. object to be selected is carried out to collision detection.Successively to C 2each candidate's object carries out collision detection, if the collision detection success finishes, otherwise, T=T+1, and forward Step3 to.
The present invention can pass through or adopt existing techniques in realizing without the technical characterictic of describing; do not repeat them here; certainly; above-mentioned explanation is not limitation of the present invention; the present invention also is not limited in above-mentioned giving an example; the variation that those skilled in the art make in essential scope of the present invention, remodeling, interpolation or replacement, also should belong to protection scope of the present invention.

Claims (4)

1. the method for a three-dimensional gesture selection object scene, comprise the steps:
(1) set up the three-dimension gesture model storehouse;
It is characterized in that, also comprise the steps:
(2) determine the desktop equation of the object of putting;
(3) initialization moment T=0;
(4) movement locus of prediction three-dimension gesture, utilize three-dimension gesture T and T+1 positional information H constantly tand H t+1, prediction straight line l:
The equation of straight line l is: x = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) x T + ( t - T ) / ( ( T + 1 ) - T ) x T + 1 y = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) y T + ( t - T ) / ( ( T + 1 ) - T ) y T + 1 z = ( t - ( T + 1 ) ) / ( T - ( T + 1 ) ) z T + ( t - T ) / ( ( T + 1 ) - T ) z T + 1 ,
T=0,1,2 ..., n, the wherein frame number of n for predicting, wherein H t=(x t, y t, z t), H t+1=(x t+1, y t+1, z t+1);
(5) tentatively selected object collection C1 to be selected;
(6) finally determine object collection C2 to be selected;
(7) show object to be selected, the candidate's object in set C2 is carried out to high light display and show;
(8) object to be selected is carried out to collision detection, successively each candidate's object of object collection C2 is carried out to collision detection, if the collision detection success finishes, otherwise, T=T+1, T is constantly, goes to step (4).
2. three-dimensional gesture according to claim 1 is selected the method for object scene, it is characterized in that, described step (2) comprises the steps:
(2.1) select the three-dimensional coordinate of known desktop lower-left angle point A in three-dimensional scenic, and the known vector that intersects at Liang Tiao limit, the desktop lower left corner, i.e. two crossing vectors on desktop
Figure FDA0000385789350000012
known;
(2.2) by an A, vector
Figure FDA0000385789350000021
obtain the plane equation of desktop: Z=f (x, y, z).
3. three-dimensional gesture according to claim 1 is selected the method for object scene, it is characterized in that, described step (5) comprising:
(5.1) setting threshold value to be chosen is R, wherein threshold value
Figure FDA0000385789350000022
m ifor the coordinate of object i, the number of the last object of N desktop;
(5.2) obtain an intersection point o by straight line l and desktop Z;
(5.3) calculate successively the distance of each object to intersection point o: r ithe object of<R is candidate's object, joins the set collection C of candidate's object 1in.
4. three-dimensional gesture according to claim 1 is selected the method for object scene, it is characterized in that, described step (6) comprising:
(6.1) calculated candidate object C successively 1the cosine value of the angle between three gesture current locations of set and the normal vector of the formed vector of each candidate's object and hand; Cos<n, H t+1-M i, the normal vector that n is hand, the normal vector of hand is the vector perpendicular to palmar aspect;
(6.2) judgement cos<n, H t+1-M iwhether be greater than 0, if be greater than 0, this object is candidate's object, joins candidate's object collection C 2in, this step is exactly that object is got rid of at candidate's object of three-dimension gesture back.
CN201310445376.9A 2013-09-23 2013-09-23 A kind of three-dimensional virtual gesture selects the method for object scene Active CN103472923B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310445376.9A CN103472923B (en) 2013-09-23 2013-09-23 A kind of three-dimensional virtual gesture selects the method for object scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310445376.9A CN103472923B (en) 2013-09-23 2013-09-23 A kind of three-dimensional virtual gesture selects the method for object scene

Publications (2)

Publication Number Publication Date
CN103472923A true CN103472923A (en) 2013-12-25
CN103472923B CN103472923B (en) 2016-04-06

Family

ID=49797806

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310445376.9A Active CN103472923B (en) 2013-09-23 2013-09-23 A kind of three-dimensional virtual gesture selects the method for object scene

Country Status (1)

Country Link
CN (1) CN103472923B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105929947A (en) * 2016-04-15 2016-09-07 济南大学 Scene situation perception based man-machine interaction method
CN109271023A (en) * 2018-08-29 2019-01-25 浙江大学 A kind of selection method based on three dimensional object appearance profile free hand gestures manual expression
CN110837326A (en) * 2019-10-24 2020-02-25 浙江大学 Three-dimensional target selection method based on object attribute progressive expression
CN114366295A (en) * 2021-12-31 2022-04-19 杭州脉流科技有限公司 Microcatheter path generation method, shaping method of shaped needle, computer device, readable storage medium and program product

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050166163A1 (en) * 2004-01-23 2005-07-28 Chang Nelson L.A. Systems and methods of interfacing with a machine
US20080030460A1 (en) * 2000-07-24 2008-02-07 Gesturetek, Inc. Video-based image control system
CN101952818A (en) * 2007-09-14 2011-01-19 智慧投资控股67有限责任公司 Processing based on the user interactions of attitude
CN102270275A (en) * 2010-06-04 2011-12-07 汤姆森特许公司 Method for selection of an object in a virtual environment

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080030460A1 (en) * 2000-07-24 2008-02-07 Gesturetek, Inc. Video-based image control system
US20050166163A1 (en) * 2004-01-23 2005-07-28 Chang Nelson L.A. Systems and methods of interfacing with a machine
CN101952818A (en) * 2007-09-14 2011-01-19 智慧投资控股67有限责任公司 Processing based on the user interactions of attitude
CN102270275A (en) * 2010-06-04 2011-12-07 汤姆森特许公司 Method for selection of an object in a virtual environment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
EIICHI HOSOYA等: "Arm-Pointer:3D Pointing Interface for Real-World Interaction", 《COMPUTER VISION IN HUMAN-COMPUTER INTERACTION》, 13 May 2004 (2004-05-13), pages 72 - 82, XP019006110 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105929947A (en) * 2016-04-15 2016-09-07 济南大学 Scene situation perception based man-machine interaction method
CN109271023A (en) * 2018-08-29 2019-01-25 浙江大学 A kind of selection method based on three dimensional object appearance profile free hand gestures manual expression
CN109271023B (en) * 2018-08-29 2020-09-01 浙江大学 Selection method based on three-dimensional object outline free-hand gesture action expression
CN110837326A (en) * 2019-10-24 2020-02-25 浙江大学 Three-dimensional target selection method based on object attribute progressive expression
CN110837326B (en) * 2019-10-24 2021-08-10 浙江大学 Three-dimensional target selection method based on object attribute progressive expression
CN114366295A (en) * 2021-12-31 2022-04-19 杭州脉流科技有限公司 Microcatheter path generation method, shaping method of shaped needle, computer device, readable storage medium and program product

Also Published As

Publication number Publication date
CN103472923B (en) 2016-04-06

Similar Documents

Publication Publication Date Title
US11875012B2 (en) Throwable interface for augmented reality and virtual reality environments
Hackenberg et al. Lightweight palm and finger tracking for real-time 3D gesture control
JP6839778B2 (en) Gesture detection methods and devices on the user-referenced spatial coordinate system
Rabbi et al. A survey on augmented reality challenges and tracking
US20170153713A1 (en) Head mounted display device and control method
US10120526B2 (en) Volumetric image display device and method of providing user interface using visual indicator
JP2019519387A (en) Visualization of Augmented Reality Robot System
US9361665B2 (en) Methods and systems for viewing a three-dimensional (3D) virtual object
CN103793060A (en) User interaction system and method
US20130215109A1 (en) Designating Real World Locations for Virtual World Control
CN104317391A (en) Stereoscopic vision-based three-dimensional palm posture recognition interactive method and system
CN102609942A (en) Mobile camera localization using depth maps
CN102236414A (en) Picture operation method and system in three-dimensional display space
CN103472923A (en) Method for selecting scene objects with three-dimensional virtual gesture
US20180113596A1 (en) Interface for positioning an object in three-dimensional graphical space
CN103677240A (en) Virtual touch interaction method and equipment
Renner et al. A path-based attention guiding technique for assembly environments with target occlusions
CN115335894A (en) System and method for virtual and augmented reality
US9870119B2 (en) Computing apparatus and method for providing three-dimensional (3D) interaction
US20220379473A1 (en) Trajectory plan generation device, trajectory plan generation method, and trajectory plan generation program
JP2004265222A (en) Interface method, system, and program
KR101338958B1 (en) system and method for moving virtual object tridimentionally in multi touchable terminal
Wang et al. Disocclusion headlight for selection assistance in vr
Rodriguez et al. Robust vision-based hand tracking using single camera for ubiquitous 3D gesture interaction
Lee et al. Tunnelslice: Freehand subspace acquisition using an egocentric tunnel for wearable augmented reality

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant