WO2008081961A1 - 情報処理方法 - Google Patents
情報処理方法 Download PDFInfo
- Publication number
- WO2008081961A1 WO2008081961A1 PCT/JP2007/075356 JP2007075356W WO2008081961A1 WO 2008081961 A1 WO2008081961 A1 WO 2008081961A1 JP 2007075356 W JP2007075356 W JP 2007075356W WO 2008081961 A1 WO2008081961 A1 WO 2008081961A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- processing method
- information processing
- area
- rectangular
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/10—Geometric effects
- G06T15/20—Perspective computation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
Definitions
- the present invention relates to a method for displaying information included in a continuous surface when information is displayed on a rectangular plane, for example.
- Information that reduces local distortion when displaying information on the earth surface as a typical example It relates to the processing method.
- IPIX Non-patent Document 1
- Non-patent Document 1 Non-patent Document 1
- IPIX Non-patent Document 1
- the cylindrical projection method of the world map creation method projects the spherical surface onto a cylinder in contact with the equator and expands it to flatten the sphere into a rectangular shape and display it as a whole.
- the Mercator projection is a regular projection
- the cylindrical product projection is a positive product projection.
- a similar technique is a panorama photo in which 360 degrees of the entire circumference are divided into multiple images and the images are pasted on a virtual cylindrical surface.
- Non-Patent Document 2 published in 1879 by Charles Sanders Peirce, the US Coast and Geodetic Survey can provide a square and 1: 2 rectangular conformal world map. Multiple maps can be placed side-by-side. Geographic information between adjacent world maps is the same. (Hereafter referred to as “Peirce's projectionj”.)
- Non-Patent Document 4 Backminster's Fuller's Dimaxion Map (Non-Patent Document 4) is used to correct the distortion of spherical information that is flat and to project and expand spherical information onto an icosahedron.
- Each equilateral triangle surface of the icosahedral spherical information corresponds to the whole sphere of each surface.
- the area and shape of the continent are flattened so that the center angle of each side of the equilateral triangle is 63 degrees 26 minutes when projecting all spherical information onto an icosahedron.
- the projection method is less than the conventional one.
- Non-Patent Document 6 is a diversion of the world map projection method to image processing of photographs.
- planar images such as panoramic photographs and photographs taken with all-around fisheye are input, it is Conversion based on many world map projection methods including imaxion map (Non-patent document 4), quincuncial projection, (Non-patent document 2), etc., and conversion into polyhedral images can be performed.
- Patent Document 1 discloses a technique for combining omnidirectional images using a technique called a mesh camera in a technique for pasting photographs taken with different types of lenses without deviation. Projection can be performed on a solid such as a regular polyhedron or a sphere called an output format while minimizing the change of the above.
- Immersive imaging method and apparatus (hereinafter referred to as “*, I Immersive patent”) (Patent Literature 2) is a regular image with the optical axis outwardly arranged on 11 of the dodecahedron surfaces.
- Optical axis setting based on regular tetrahedrons and regular octahedrons has been proposed that can simultaneously capture substantially omnidirectional images of dodecahedrons and take stereoscopic vision into account.
- Non-Patent Document 1 The spherical information displayed by IPIX (Non-Patent Document 1) is greatly distorted when the viewer is zoomed out, and the image cannot be recognized in the field of view equivalent to a hemisphere. Also, because the setting is along one axis, the movement of the viewer becomes unnatural, and a gimbal lock and rebound phenomenon occurs.
- Cylindrical projection is a planar display method in which distortion concentrates on the upper and lower parts of an image, and it is difficult to recognize the objects at the upper and lower ends.
- geographical information can only be rearranged in the east-west direction without complicated image processing.
- the Dimaxion 'map (Non-Patent Document 4) is an expanded view of an icosahedron, so the outer shape of the world map is zigzag and it is difficult to recognize the geographical relationship.
- the ocean current is expressed in this, even if the 20 equilateral triangles of the map are rearranged, the continuous relationship of the ocean is broken. In other words, it is impossible to fill geographic information without gaps in an ideal rectangular plane as a map.
- the icosahedron used by the dimaxion 'map is a regular polyhedron that can divide the omnidirectional image into 20 equal parts, and has the least amount of distortion in the regular polyhedron.
- the object near the apex of the regular tetrahedron, which only divides the omnidirectional image into four equal parts and has large distortion, is enlarged to more than five times the actual solid angle.
- This technology is not a proposal for rectangular planar images.
- Peirce's projection (Non-Patent Document 2) is also distorted for the same reason as above.
- Collumble's Projection (Edouard Collignon, 1865) of Non-Patent Document 5 is an equal area projection, and the outline of the map is simple but is not a proposal to fit in a rectangle.
- Non-patent Document 6 which diverted the world map projection to a photograph, has the above-mentioned problems, like Dimaxion, Map (Non-patent Document 4) and Peirce's projection (Non-patent Document 2). Yes.
- Patent Document 1 is not an omnidirectional image planarization technique.
- the imaging method there is no image, so in the imaging area, that is, in the case of imaging in which the solid angles of each imaging device are arranged unevenly, an image area that is locally extended when the image is made into a regular polyhedron is created, The resolution varies.
- Patent Document 2 is also not a plane image technique for omnidirectional images.
- the optical axis is not set on one surface of the regular dodecahedron, and it is not complete omnidirectional information.
- Patent Document 1 JP-A-2003-178298 (“Image Processing Apparatus and Image Processing Method, Storage Medium, and Computer Program”, Sony Corporation)
- Patent Document 2 US Patent No. 6141034 (("Immersive imaging method and apparatus], Immersive media Co.)
- Non-Patent Document 1 IPIX IPIX 360 SuiteJ (http://www.ipix.com/)
- Non-Patent Document 2 Quincuncial projection, Charles Sanders Peirce, the U.S. Coast and
- Non-Patent Literature 3 riangukr projection, Laurence P. Lee, national mapping agency in New Zealand, 1965 (http://www.progonos.com/fliruti/index.html)
- Non-Patent Document 4 Dymaxion MapJlNVENTIONSj, R. Buckrainster Fuller (St. Martins'
- Patent Literature 5 Collumble's Projection, Edouard Collignon, 1865 (http: / /www.progon os.com/ furuti / index, html)
- Patent Literature 6 Flexy2, flaming pear (http://www.flamingpear.com/flexify.html) Disclosure of Invention
- An object of the present invention is to provide an information processing method capable of reducing local distortions when, for example, information on the earth is developed on a plane or vice versa. .
- a further object of the present invention is to provide an information processing method capable of displaying on a rectangular plane while maintaining the area ratio of continents and islands on the earth, for example.
- the invention of the present application is typically characterized in that information is generated in a rectangular plane by repeating surface division or surface integration using a spherical polyhedron that is inscribed or circumscribed.
- information is generated in a rectangular plane by repeating surface division or surface integration using a spherical polyhedron that is inscribed or circumscribed.
- this is explained as a superordinate concept, it is characterized in that the area ratio of each surface is maintained with respect to all the areas to be divided or integrated.
- the distortion of information can be reduced and / or the distortion can be uniformly distributed.
- each surface maintains the relative positional relationship of each surface in a plurality of start surfaces with continuous surfaces defined by lines and points, maintain the lines of each surface as lines, and maintain the points of each surface as points.
- it is an information processing method in which each surface is deformed and filled in the rectangular plane without gaps, or vice versa, so that the information on the start surface and the information on the plurality of end faces that fill the rectangular plane without gaps correspond one-to-one.
- the first area ratio of the total area of the start surface and the area of the start surface, and the second area ratio of the area of the rectangular plane and the area of each end surface on the rectangular plane are substantially
- An information processing method characterized by equality is provided. This is achieved by reducing distortion and / or distributing uniformly.
- a plurality of starting surfaces defined by lines and points are consecutively defined.
- FIG. 60 illustrates the case where two photographs are taken with a fisheye lens all around, they are defined by lines SE1 and SP1.
- the surface SS1 and the surface SS2 defined by the line S E2 and the point SP1 are shared by one point SP1.
- each surface and maintaining the lines of each surface as lines and maintaining the points of each surface as points deforming each surface and filling it in a rectangular plane without gaps
- the two surfaces deformed and filled in the rectangular plane without gaps are the surfaces RS1 and RS2, but the two surfaces share one point RP1 and maintain a continuous positional relationship, and the point SP1 is the line RP1 and the line SE1 is RE1 and SE2 are replaced by RE2, and the surface SS1 and surface SS2 are deformed to surface RS1 and surface RS2 while maintaining the point of each surface as a line with the line of each surface before deformation as a line, and there is no gap in the rectangular plane Represents filling.
- the information on the start surface and the information on the plurality of end surfaces that fill the rectangular plane without gaps are in one-to-one correspondence.
- the information on the surface SS10 and the information on the surface RS10 are in one-to-one correspondence. It means that Further, the first area ratio of the total area of the start surface and the area of the start surface and the second area ratio of the area of the rectangular plane and the area of each end surface on the rectangular plane are substantially Is equal to
- the present invention proposes an information processing method that fills a rectangular plane without any gap while satisfying the above conditions.
- the surface PS100 is composed of, for example, three lines PE12 to PE14.
- the surface PS11 defined by the points PP12 to 14 is continuous with the surface PS12 sharing the line PE12, the surface PS13 sharing the line PE13, and the surface PS14 sharing the line PE14.
- RS100 is obtained.
- Lines PE12 to 14 are replaced with lines RE12 to 14, respectively, points PP12 to 14 are replaced with points RP12 to 14 and L4, respectively.
- the line of each surface before deformation is a line, and the points of each surface are maintained as points and points, while maintaining the relative positional relationship of each surface before deformation and the information on each surface RS11-14 and surface RS11-; It shows how each of the 14 pieces of information is in a one-to-one correspondence. .
- a first area ratio of the total area of the start surface and the area of the start surface, and a second area ratio of the area of the rectangular plane and the area of each end surface on the rectangular plane; are substantially equal
- PS11 / PS100 RS11 / RS100
- a plurality of surfaces with continuous surfaces defined by lines and points includes a plurality of continuous surfaces generated by a certain operation, as shown in the example of FIG. Brief description of the drawings
- FIG. 1 is a diagram showing the idea of a solid angle.
- FIG. 2 is an explanatory diagram showing an example of a regular tetrahedron and a spherical regular tetrahedron in a regular product grid according to the present invention.
- FIG. 3 is a perspective view of the two grids of FIG.
- FIG. 4 is an explanatory diagram showing a grid obtained by optically projecting the spherical grid shown in FIG. 2 onto a regular tetrahedron.
- FIG. 5 is a perspective view of the two grids of FIG.
- FIG. 6 is an explanatory diagram showing a grid obtained by subdividing the two grids of FIG.
- Fig. 7 is an explanatory view showing the earth which is an area product mapped to a regular tetrahedron according to the present invention.
- FIG. 8 is an explanatory view showing a world map obtained by performing positive product mapping according to the present invention and a world map excluding the product map.
- FIG. 9 is an explanatory view showing a plane filling image according to the present invention showing a viewer.
- FIG. 10 shows four world maps obtained by the plane filling image force of FIG.
- FIG. 11 shows four other world maps obtained with the plane filling image force of FIG.
- FIG. 12 is an explanatory view showing a plane filling image showing an omnidirectional image unit different from FIG.
- FIG. 13 is a current chart in January of the world shown by Dymaxion Map.
- FIG. 14 is a current chart of January in the world shown by a world map according to the present invention.
- FIG. 15 is an explanatory diagram showing a cubic product and a spherical cube as an example of an equal product grid according to the present invention.
- FIG. 16 is an explanatory diagram showing a graticule equal product grid on omnidirectional spherical information.
- FIG. 17 is an explanatory view of the omnidirectional spherical information of FIG. 16 as viewed from above.
- FIG. 18 is an explanatory diagram showing a graticule grid product grid on a regular octahedron.
- FIG. 19 is an explanatory diagram in which the grid shown in FIGS. 16 and 18 is developed on a plane.
- FIG. 20 is a diagram showing a grid obtained by mapping a regular octahedron image to two square surfaces.
- FIG. 21 is an explanatory diagram showing a process of integrating omnidirectional images mapped on the front and back of a square into one square.
- FIG. 22 is an explanatory diagram showing three square world maps created from one regular octahedron image.
- FIG. 23 is an explanatory diagram showing a plane-filled image of the world map of FIG.
- FIG. 24 is an explanatory view showing a plane-filled image of another world map of FIG.
- FIG. 25 is an explanatory view showing a plane-filled image of another world map of FIG.
- FIG. 26 is a diagram showing an equal product map from an icosahedron to an icosahedron.
- FIG. 27 is an explanatory diagram showing a product mapping from a regular dodecahedron to a cube.
- FIG. 28 is a schematic diagram showing an entire background image according to the present invention.
- FIG. 29 is a schematic diagram showing a planar filling image of a negative-inverted template image.
- FIG. 30 is a schematic diagram showing a search image obtained by synthesizing the plane filling images of FIGS. 28 and 29.
- FIG. 31 is a schematic diagram showing an omnidirectional cubic image pickup device.
- FIG. 32 is a schematic diagram showing an image pickup device sharing an optical center.
- FIG. 33 is a schematic diagram for explaining details of the image pickup device sharing the optical center of FIG. 32.
- FIG. 34 is a schematic diagram of a rectangular display operation screen for mapping.
- FIG. 35 is a schematic diagram showing a process of mapping a planar image to a regular tetrahedron and a sphere.
- FIG. 36 is a schematic cross-sectional view showing a process for mapping an image on a sphere onto a subject.
- FIG. 37 is a schematic cross-sectional view showing a process of mapping a subject of arbitrary shape onto a sphere.
- FIG. 38 is a schematic diagram showing a first half process of multi-layer mapping of an arbitrary shape to an octahedron.
- FIG. 39 is a schematic diagram showing the latter half of the process of multi-layer mapping an arbitrary shape to a regular octahedron.
- FIG. 40 is a schematic diagram showing a rectangular operation screen for a spherical screen.
- FIG. 41 is a schematic diagram illustrating an imaging method in which a three-dimensional light receiving surface is arranged on each optical axis, taking a cube as an example.
- FIG. 42 is a schematic diagram showing an imaging method in which a light receiving surface is made polyhedral as an example of a regular tetrahedron.
- FIG. 43 is a conceptual diagram showing the shape and arrangement of the light receiving elements along the subdivided equal product grid.
- FIG. 44 is a schematic diagram illustrating an imaging method in which the three-dimensional light-receiving surfaces are gathered in a plane, taking a regular octahedron as an example.
- FIG. 45 is a schematic diagram showing an imaging method taking into account imaging by a plurality of units, taking a 14-sided body as an example.
- FIG. 46 is a schematic diagram showing an example of a 26-hedron imaging method for providing diversity in optical axis setting.
- FIG. 47 is a schematic diagram showing an image pickup device that can cope with resolution and stereoscopic imaging.
- FIG. 48 is a schematic diagram showing a stereoscopic imaging method and an omnidirectional imaging method by the imaging device of FIG. 47.
- FIG. 49 is a schematic diagram for explaining an omnidirectional stereoscopic imaging method using a cubic imaging device as an example.
- FIG. 50 is a schematic diagram showing the use of three combinations of the image pickup device of FIG. 47 as an example.
- FIG. 51 is a schematic diagram showing a regular tetrahedron as an example of a method of mapping a polyhedron image to a rectangular plane.
- FIG. 52 is a schematic diagram illustrating an imaging method for quickly obtaining a rectangular image using two imagers as an example.
- FIG. 53 is a schematic diagram illustrating a solid having an open surface as a process for mixing a plurality of types of mapping methods.
- FIG. 54 is a schematic diagram showing an example of rhomboid dodecahedron as a method of compiling a cylindrical projection and a rectangular image.
- FIG. 55 shows an example of a regular octahedron in which a polyhedral and a curved surface that compromises a plane are divided into equal areas. It is the shown schematic diagram.
- FIG. 56 is a schematic diagram showing an image pickup device in consideration of downsizing.
- FIG. 57 is a conceptual diagram showing a rhomboid dodecahedron image and an octahedron image obtained by the image pickup device of FIG. 41.
- FIG. 58 is a schematic diagram showing a method of dividing by a curve, taking as an example a solid whose arc is a meridian.
- FIG. 59 is a schematic diagram showing a quadratic curve as an example of a dividing method using curves.
- FIG. 60 is a conceptual diagram showing a map according to the present invention as an example of a circular image by a fisheye lens.
- FIG. 61 is a conceptual diagram showing, as an example, a developed image of a polyhedron in mapping according to the present invention.
- FIG. 62 is a schematic diagram showing a process of integrating images from dispersed light receiving surfaces into one solid.
- FIG. 63 is a schematic view showing a map according to the present invention as an example of a circular image.
- FIG. 64 is a schematic diagram showing an example of a cylindrical projection image of the mapping according to the present invention. BEST MODE FOR CARRYING OUT THE INVENTION
- the area ratio of a specific surface to the total area of a plurality of continuous surfaces will be described by taking a solid, especially a sphere as an example.
- the area ratio of the spherical surface to the total area is called “solid angle”.
- the solid angle will be described with reference to FIG.
- the solid angle is a number that represents the power of the subject 2 viewed from the center 01 of the sphere S1, and is usually expressed in steradians (sr).
- the size is expressed by the area 3 of the upper part of the sphere when a weight made of a set of half straight lines connecting from 01 to the subject 2 is cut by a unit sphere S1 with a radius 1 centered on 01.
- the solid angle of the whole space seen from 01 is 4 sr
- the solid angle of the hemisphere is 2 sr.
- the world map corresponds to the area of the continent.
- the field of view by the angle of view ⁇ must be expressed as a small circle C2 on the unit sphere! /.
- This small circle is called an image circle.
- IPIX Non-patent Document 1
- mapping is the transformation of a single surface, the transformation of multiple continuous surfaces into a single surface, or the transformation of a single surface into multiple continuous surfaces. Including operations to divide By observing and diagnosing the rectangular display using the product map, quantitative observations that are difficult for human vision are possible. In each field such as endoscopic diagnosis, map, magnetic force measurement, sky ratio calculation, etc., the spread and distribution of diseased areas, ozone hole area, magnetic flux density and distribution, and projective area of buildings can be examined.
- Rectification by image mapping of image information such as omnidirectional will be described using a regular tetrahedron.
- regular polyhedrons By classifying on the basis of regular polyhedrons, not only the solid angle but also the length of the line segment, the surface angle, and the internal angle error, which is the angle between the line segments, can be equally distributed.
- 2 and 3 are conceptual diagrams showing the grid G1 on the regular tetrahedron P10 and the corresponding grid G2 on the spherical regular tetrahedron in contact with P10.
- 4 and 5 are conceptual diagrams showing the grid G3 obtained by optically projecting G2 onto the regular tetrahedron from the optical center O10.
- G1 the vertex of the regular triangle of the regular tetrahedron and the midpoint of the opposite side are connected by a straight line that is a geodesic line.
- a grid is created in which the lengths of line segments and the internal angles of line segments are unified into several types. One of them is the shaded area 34. In this way, the regular tetrahedron is divided into 96 equal parts.
- the midpoints 30, 35, and 36 of the arc of the spherical regular tetrahedron are connected to the vertices V6, V7, and V5 by a large arc as a geodesic line. They meet at point 09.
- the spherical quadrangle 09, 35, V 6, 36 are connected to the point 28 and the midpoint and each vertex of the edge of the spherical quadrangle by a large arc to divide the area into eight equal parts.
- the other quadrilaterals can be divided in the same way to create a grid that is equally divided into 24 types of two spherical triangles (including some mirror images), with a uniform line segment length and interior angle. One of them is region 37.
- the same operation is performed on each surface of the spherical regular tetrahedron, and the spherical surface is divided into 96 equal parts. Of course, this may be considered as a spherical 96-hedron.
- a grid G3 is obtained.
- Each grid point of grid G3 is on a straight line connecting grid G2 and center O10.
- spherical triangular regions 37 and 70a with the same solid angle (area ratio to the entire sphere) on grid G2 is 5 to 1 times the value of the area force 70b to the total area of the regular tetrahedron in the region 37a.
- the product mapping is to map to region 34 on grid G1 while maintaining the solid angle of region 37.
- the above division is called an equal product division, and the above divided grid is called an equal product grid.
- the technical matters described below are not limited to the first embodiment, but can be applied to other embodiments in the same way. It is not necessary to use geodesic lines. This includes curves that include small circles, and line segments that link these. The division may be partially uneven. It should be noted that the force S that mentions a sphere or regular tetrahedron, and the mapping in the present invention targets any solid.
- the arbitrary solid mentioned here includes a polyhedron including a regular polyhedron and a semi-regular polyhedron, a solid with open surfaces such as a hyperboloid, a solid including a curved surface, and a rotating body.
- the above three-dimensional object such as a hemisphere used for measuring the sky ratio may be partially used.
- this square product division theoretically includes the case of using only regular tetrahedron edges, but it should be understood that in applications such as surveillance cameras, it is better to use polyhedron edges that do not contain regular tetrahedrons. Yes.
- the accuracy of the equal product map can be adjusted by adjusting the number of divisions of the grid.
- the grid G5 in Fig. 6 is an 8-divided grid of spherical regular tetrahedrons obtained by subdividing G • 2. Based on this, the spherical information is mapped onto an 8-divided grid G4 of regular tetrahedrons subdividing G1. For example, the triangle 71 on the spherical surface is mapped to the triangle 71a.
- FIG. 7 is a regular tetrahedron image obtained by normal product mapping of the earth (omnidirectional information).
- the line segment connecting midpoint 9 and vertices 10 and 11 of regular tetrahedron edge E3 is E4 and E5. Let vertices at both ends of E3 be 15,16.
- This regular tetrahedron image is expanded by cutting into E4, E5, and E3, and flattened.
- point 9 in FIG. 7 is placed at each vertex 17 of the rectangle in FIG. 8, and points 10, 15, and 11 are placed at points 20, 18, and 19, respectively.
- the sum of the internal angles of the faces at each vertex is 180 degrees, so a rectangular plane is obtained in this way.
- the world map SCla is shown in which the same area map is omitted from the same process. Since the four vertices are evenly distributed and the solid and central angles of the large frame are both maintained by the regular tetrahedron uniformity, the distortion does not concentrate locally.
- any flat line on the regular tetrahedron may be cut into a rectangular plane. If you are not particular about the rectangle, you may develop it into a shape that includes a polygon such as a regular triangle or a curve such as a circle. It can be any 3D surface, such as a curved surface. A development of a regular tetrahedron is preferable, but other polyhedrons may be used as long as they do not stick to a rectangle.
- a dotted line E 7 is a line segment in which a regular tetrahedron ridge is mapped, and forms a three-way grid.
- Viewers VR1, VR3, and VR4 can be obtained for one world map that has a different aspect ratio than the world map SC 1 described above. They can change direction and move continuously in three directions along the three-way grid 23, 24, 25.
- SC1 is a world map centered on Antarctica and Australia, but it can provide an appropriate omnidirectional image of a secondary cell and an equivalent world map by moving the viewer for one world map.
- Fig. 10 and Fig. 11 show world maps with an aspect ratio of 4: ⁇ 3 centered on various regions where the plane filling image power can be obtained. These are world maps that show the continent without any substantial division.
- LC1 ⁇ 8 are the world-wide world maps centered on the Middle East, Pacific Ocean, Antarctica, India and China, Europe, Central America, Oceania and Japan, respectively.
- FIG. 12 shows that the same plane-filled image as FIG.
- the rectangular world maps SC2, SC3, and SC4 having an aspect ratio of 1: 3 can be extracted in addition to the above-described map.
- the SC40 world map SC40 can be provided so that the geography around the four corners of the map SC4 (near Zealand) can be grasped.
- 16 ⁇ 3 ⁇ Viewer SC400 can grasp the geographical relationship with the whole world at the four corners of the map. These can show the currents and routes of the world, meteorological observation maps, and the locus of satellites without interruption.
- equilateral triangles other than the rectangular cell of the first embodiment described above. You may take out a polygon such as a cell. Northern Hemisphere etc. 4 Tc sr or less may be taken out. In this example, the regular tetrahedron development is filled horizontally and vertically, but it may be replaced with another polyhedron. They may be arranged in a single row or a ring. A planar arrangement with a gap may be used. When filling a flat surface, there is no need to display some subjects continuously.
- the world map is viewed from all directions toward the center of the subject called the Earth, but an omnidirectional photograph that views all directions outward from one viewpoint is also available. Regardless of whether the optical axis is inward or outward, the omnidirectional information can be treated as a spherical image. Conventional technologies treat these technologies as separate objects, but the present invention treats them equally. Therefore, the force that can be similarly applied to other embodiments may be applied to the omnidirectional photographic technique.
- FIG. 15 shows only the cubic product grid G6 and the spherical product grid G7 that shares the vertices and circumscribes it.
- Grid G6 is an equal product grid that draws a diagonal line of a square surface, connects the midpoints of each side, and equally divides it into 16 triangles. One of them is region 82. This divides the cube into 96 equal parts.
- Grid G7 connects the diagonals of the spherical square with a large arc and connects intersection 013a with each vertex with a large arc. Furthermore, the point 013a and each vertex A grid that equally divides 16 spherical triangles by connecting the midpoints 78-81 of the large arc connecting the midpoints of the arcs and the midpoints of each side of the spherical square with a large arc. One of them is region 82a. In this way, the spherical surface (spherical cube) is divided into 96 equal parts. Thus, for example, the area 82a on the spherical surface can be mapped to the area 82 on the cube.
- the cubic image obtained by the orthogonal product mapping is orthographically projected onto each surface of the regular tetrahedron having points V8, V9, V12, and V13 as vertices, resulting in an orthogonal tetrahedron. If the same mapping is performed on the regular tetrahedron with the vertices V10, VII, V14, and V15, another regular tetrahedral image can be obtained.
- the equal area division of the present invention includes an equal area grid in which solid angles are equally divided and then further subdivided with graticules or the like.
- the normal product map of the present invention includes a method of using a polyhedron having a rectangular cross section to convert the image information of the polyhedron into a rectangular flat surface by mapping the polyhedron image information onto the cross section.
- the spherical octahedron is divided into equal parts by dividing the area ratio while maintaining a rough central angle by the ridgeline, and then subdividing along the graticule into a regular product grid and a corresponding graticule product grid on the octahedron.
- An embodiment will be described in which a square product, in particular a square, omnidirectional image is provided by mapping the product onto the front and back of the rectangle after performing the normal product mapping.
- the earth which is omnidirectional spherical information shown in Fig. 16, is equally divided by the ridgelines 90, 91, 92 of the spherical regular octahedron.
- 90 is the equator and 91 and 92 are meridians.
- Grid G12 subdivides a spherical regular octahedron with graticules. One of them is area 89.
- a part of grid G12 is shown on sphere S6.
- the meridians are arranged so that the angles between the meridians are equal, and the parallels are parallel to each other.
- the intersections 93 and 94 of meridians 92 and 91 correspond to the north and south poles.
- Figure 17 shows S6 from the top, that is, the North Pole 93.
- FIG. 18 is a part of the grid G9 on the regular octahedron P5 that shares and touches the sphere S6, and the squares 95, 96, and 97 share the vertex with this regular octahedron.
- Grid G9 is composed of meridians connecting points 102, 103, and 104 that bisect the north pole 93 and the vertices of the regular octahedron, and parallels parallel to the lines connecting 98 and 99. You may arrange
- the region 89a is a one-segment region by G9. In this way, each segmented area is mapped from grid G12 to G9, for example, area 89 is mapped to area 89a.
- a grid Gl 1 shown in FIG. 19 is a flat development of the grid G12 of FIG.
- Grid G 9 is a front view of the grid on the regular octahedron P5.
- FIG. 21 is a perspective view of a regular octahedron P5.
- the images obtained in this way are, for example, the world maps LC13, LC14, and LC15 shown in Fig. 22.
- the Mercator projection displays distorted Antarctic continents and Arctic sea ice fields in the same product, and concentric parallel parallels and polar forces. Relationship is maintained.
- Figures 23 and 24 show the above-mentioned images LC14 and LC15 arranged as a cell (unit) image, and are filled in the plane.
- the ice fields in Antarctica and the Arctic Ocean are located near the midpoints 140 and 141 of each side of the square. However, there are rare cases where two are connected continuously.
- the LC13 filled with plane is shown in Fig. 25, and the whole land can be displayed comfortably. In this way, three omnidirectional images can be obtained from one regular octahedron, and an easy-to-see horizontal plane filling image can be selected.
- the secondary cell LC20 centered at the North Pole surrounded by points 122, 133, 134, and 132 can be extracted from the planar filling image shown in FIG. 25, and a world map with an aspect ratio of 1: 2 can be extracted.
- Cell LC16, surrounded by points 121,129,130,131, is an example of this, which displays the North Pole accurately, and makes it easy to see subjects in the Southern Hemisphere that are distorted by Lambert's equal product projection.
- the world map LC17 with an aspect ratio of 1: 4 can also be taken out.
- This area LC17 can be a viewer that continuously moves and changes direction in the direction of arrow 113, and a rectangular world map LC18 can be taken out.
- This area LC18 can also be continuously moved and changed directions in the direction of arrow 114, and LC16 can also be made a viewer that is continuously moved and changed in direction of arrow 115, and the rectangular world map LC19 can be taken out.
- LC19 can also be a viewer that continuously moves and changes direction in the direction of arrow 116. In this way, a viewer with a maximum 4 ⁇ sr region with a selectable aspect ratio can be set, and a rectangular world map centered on any region can be created.
- the technical matters described below are not limited to the third embodiment, but can be applied to other embodiments as well.
- the viewer may be deformed while moving.
- the viewer should have an arbitrary shape different from the rectangular shape with the above aspect ratio.
- the positive integral division may be approximated or simplified in consideration of the processing power of the computer. In this case, the strain dispersion is limited.
- mapping 3D to a plane the process of discarding distance information and importing it into the unit sphere may be omitted. In this case, it is an operation to deform the 3D space.
- the celestial sphere information is converted while maintaining the distance information, a certain three-dimensional star space can be seen from a rectangular image, and if the viewpoint is changed, a special three-dimensional space that can grasp the distance to each celestial body can be obtained.
- the target space is cut out in layers concentrically around a point that is a sphere with various radii, large and small, and the subject between the spherical surfaces is divided and mapped to each sphere, and then the present invention is applied. Rectangular plane based Or z-plane arrangement. In this case, if the plane arrangement is performed in the order of the radius, the distance information from the center point can also be indicated by the plane coordinates depending on the arrangement position.
- FIG. 51 is a conceptual diagram of an omnidirectional tetrahedral image PG16. Map two regular tetrahedral faces of the regular tetrahedron to the square F23. As long as F23 is parallel to the ridgeline of the regular tetrahedron, the inside and outside of the regular tetrahedron PG16 may be used. The other two equilateral triangular faces are also mapped to the back of F23. An omnidirectional image can be obtained by rotating the back side and integrating it with the front side image. Further, these can be used and filled in the same plane as in the above embodiment.
- a regular polygon such as a regular hexagonal image may be obtained by mapping a cube or the like from an oblique direction and arranged in a plane. Moreover, it may be mapped to a polygonal surface of a polyhedron that forms a great circle by connecting lines such as a quasi-regular 32hedron. In this case, two images are obtained. If these are arranged while maintaining the continuity of the subject, a flat array image with gaps is formed. The mapping may be just one of these surfaces. Remap the regular tetrahedron image to the regular octahedron image PG17.
- the plane-filled images handled by the present invention include those given time characteristics (hereinafter referred to as all background images).
- the plane-filled image MX1 in FIG. 28 is an explanatory diagram of a plane-filled image in which one omnidirectional rectangular image SC20 is taken as a cell and one image is taken per second and arranged in 60 horizontal rows and 60 vertical rows in time series. You can get an image of the entire history of one hour in which time and space are continuous in seconds in the horizontal direction and minutes in the vertical direction.
- each cell is projected in time series using the above-mentioned viewer, it can be displayed like a moving image.
- the moving image recording can be provided as a single image in which the continuity of the subject is ensured in the boundary area of each cell.
- the entire background screen of the fifth embodiment is not limited to each cell. Replace still images with videos. For example, if a 1-second movie is displayed in each cell according to the imaging interval, the entire history can be viewed in a 1-second movie. It may be arranged in a three-dimensional arrangement, such as overlaying 25 frames of images per second in the depth direction of the display screen. Depending on the monitoring target, the imaging frequency, shirt speed, number of vertical and horizontal frames, etc. may be arbitrary. If you take one image per minute and arrange 60 images horizontally by 24 images vertically, you can see the events of the day. it can.
- the cells When installed on a hustle and bustle, if you set the shatter speed to 1 minute, you will be able to see only the suspicious objects that are left unattended.
- the cells may be arranged vertically or horizontally, leaving gaps, or arranged in a single row or ring. It is also possible to arrange each cell in a non-time series. It is good even if the image in each cell is partially replaced.
- the present invention includes a search image for extracting only an abnormal part from the all background images.
- Image MX2 in Fig. 29 is an explanatory diagram of a plane-filled image in which cell image SC20a is repeated 60 frames vertically and 60 frames.
- the image SC20a is a negative inversion of the above-described omnidirectional image SC20 and is used as a template image representing the normal time of the monitoring space.
- FIG. 30 is an explanatory diagram of the search image MX3.
- the image MX3 is a composite of the time-series plane-filled image MX1 and MX2 obtained by repeating the template image.
- the solid white area in Fig. MX3 is the unchanged area. This is because the subject captured in the template image and the subject imaged during monitoring are complementary and have a solid color when there is no change.
- the subject 61 that has changed can lose its complementary color relationship and be displayed as a symbol, making the subject 61 stand out as an event during the monitoring period. In this way, the image area around the subject 61 can be quickly enlarged and verified, and the context can be examined while driving the viewer.
- the unchanged part is plain, and the capacity can be greatly compressed using an image compression technique.
- the original image can be obtained by synthesizing the image SC20 separately recorded on the plane filling image. In this way, it is possible to reduce the capacity of plane filling image data in which the number of pixels rapidly increases during long-time recording.
- the template image is arbitrarily updated. For example, every time the number of cells at a certain interval or at random, the template images at different times are mixed, or the images at different times are also mixed in the divided area within the template image.
- the background of all background images may be shifted and reversed, and for example, the negative of the image shifted by one frame and the original image may be combined to display a short time change. Also, the entire background image may be negatively inverted.
- Solid color or single color image by synthesizing one or both color elements such as RGB, CMYK, HSB, Lab, etc. of the image to be synthesized, or adjusting the transparency to equalize the saturation, brightness, hue, brightness, etc. For example, the same effect may be obtained by making the change part stand out.
- the mapping in the present invention includes a plane-to-plane mapping.
- Figure 63 shows how to round a circular image with an all-around fisheye lens.
- Region SS200 is an array of all-around fisheye images that are captured in two directions in all directions.
- a grid based on the line segment SE20 is created based on the solid angle at the time of imaging, while focusing on the projection characteristics of the lens based on the equal product division by the grid G2 shown in FIG.
- a part of line segment SE20 becomes a curve.
- Region SS200 is equally divided into 96 image regions SS20. The hatched area is one of them.
- Rectangular plane RS200 is divided by a grid with line segment RE20. Shows one of the line segments RE20.
- line segments RE20 are equally divided into 96 image regions RS20 based on the equal product division by the grid G1 shown in FIG. The hatched area is one of them.
- mapping each region SS20 to the corresponding region RS20 the entire SS200 is mapped to the RS200 and squared.
- FIG. 64 shows a method for rectangularizing the plane information by the conventional projection such as an existing world map based on the present invention.
- Region CR10 is a conceptual diagram of conventional cylindrical projection that projects spherical information, typically the Earth SPR10.
- the lines SPE1-8 on the spherical surface SPR10 are projected onto the lines CE1-8, and the areas SPR1-8 are projected onto the areas CR1-8, respectively.
- Lines CE1-8 divide region CR10 into eight equal regions CR1-8.
- the region OR10 is a rectangular planar region based on the method illustrated in FIG. Line segments ⁇ 1-8 divide region OR10 into eight regions ORl-8.
- the region OR10 typically shows a world map, and because it shows the geographic information of one earth without any excess, it is natural that the geographical information around the point 0-S located at the four corners of the region OR10 is difficult to grasp. . This is because a spherical surface like the Earth originally has an omnidirectional spread at an arbitrary point on the surface as previously understood as an infinite plane. It can be said that the relative geographical relationship can be completely reproduced.
- the regions ORl to 8 can be repeatedly arranged around the region OR10 as necessary. For example, looking at such an array region OR100, it can be seen that the relative geographical relationship of the regions OR5 to 8 facing the outer periphery of the region OR10 is maintained. For example, on the sphere, the region SPR8 is adjacent to SPR4,5,7 via a line, and the OR8 indicated by the hatched region is adjacent to OR4,5,7 via a line.
- FIG. 26 is an explanatory diagram of the mapping to the icosahedron.
- a regular dodecahedron is obtained by replacing the vertex of a regular icosahedron with a face and the face with a vertex.
- the center 39 of the regular triangle region of the omnidirectional regular icosahedron image, the vertex 38, and the midpoint of the side are connected and divided into six regions, for example, the region 72.
- This region is mapped to a region divided into 120 equal parts by a line segment connecting the center 4 of the regular dodecahedron that touches the regular icosahedron to the midpoint of the side and the vertex 39 of the regular dodecahedron.
- the region 72 is mapped to the region 73 on the regular dodecahedron.
- FIG. 27 is a conceptual diagram of mapping from a regular dodecahedron to a cube.
- the side 41 of the cube in contact with the regular pentahedron of the dodecahedron divides each face of the dodecahedron into a triangle 46 and a trapezoid 48.
- This region is mapped to a trapezoid 48a having points 49 &, 42, 43 on the cube at the vertices and triangles 46 & and 4950 &, 44, 43 at the vertices, respectively.
- the positions of the points 49a and 50a may be adjusted so that the map becomes a product product map.
- the resulting cube image is a regular tetrahedron or a regular octahedron. You can create a rectangular image by mapping it back to the body.
- a multi-layered map based on a product map can be formed by combining symmetric and geometrical objects such as a regular polyhedron.
- Multi-layered mappings from geodesic spheres to quasi-regular 32-hedrons, regular dodecahedrons, and regular icosahedrons are also equivalent products.
- the present invention includes a mapping to a plane force solid in the reverse process of the above mapping.
- CG computer graphics
- a texture image of a rectangular plane is created, and texture mapping with less distortion can be performed by mapping an equal product onto the surface of the solid.
- An image for the entire three-dimensional surface is drawn in the rectangular image area TM1 in which the equal product grid shown in FIG. 34 is arranged.
- a triangle 147 is a part of TM1 and shows an image area in which a product grid represented by dotted lines in the figure divided based on the division shown in the first embodiment is arranged. Other areas are similarly divided.
- An image area TM2 larger than 4 ⁇ ⁇ is used here so that the image displayed in the area can be repeatedly arranged around TM1 and the joint between the ends of TM1 can be seen.
- the image in the area 145 is also displayed in the area 145a.
- FIG. 36 is a cross-section taken along the great circle 159, and is a schematic diagram showing the positional relationship between the objects obj6 and obj5. With the center 160 of obj5 as the center of light, the image on obj5 is mapped to obj6. Obj5 may be placed outside or crossing the subject.
- the present invention includes spherical surface information obtained by setting an optical axis of a subject having an arbitrary shape inward in all directions.
- Fig. 37 shows a cross section of the object objl in the shape of a human head.
- the entire surface of the subject is divided and imaged by a plurality of imaging devices, and the spherical surface information is obtained by mapping the spherical surface 162 by setting the optical axis inward from all directions.
- the spherical information is converted into a rectangular planar image as in the planarization technique described above.
- objl maps directly to a regular tetrahedron without going through a sphere.
- the present invention includes a multi-layered mapping of an arbitrarily shaped solid.
- An example of mapping a substantially full surface image of an arbitrary shape to a regular octahedron by multi-layer mapping will be described.
- Figure 38 shows the subject (person's head) obj3.
- a region obtained by dividing the total surface area into eight equal parts by dotted lines 155 and 156 is the shaded area AR1.
- AR1 is a partially enlarged drawing power AR2.
- AR2 is a polygonal cone consisting of polygons SR4, SR5, and SR6 on obj3.
- This operation is performed at each location of obj3 to obtain a solid obj2 with fewer polygons in FIG.
- a plurality of such surfaces are repeatedly integrated into one surface by an equimature map.
- many polygons of AR1 are integrated into one equilateral triangle and integrated into an octahedral image obj7.
- the regular octahedron image is further plane-integrated to obtain a square image.
- it is suitable when the surface of a complex shape that can form a blind spot with a single optical center is formed into a rectangular flat surface by gradually simplifying and arranging the shape gradually by multi-layer mapping.
- the present invention includes a display method capable of manipulating images while overlooking a omnidirectional image to be input / output on a rectangular plane. For example, when projecting image material that captures omnidirectional images onto a spherical screen, multiple projectors and imagers are required for input and output, but the solid angle of the projection area and imaging area is incorrect on the conventional display screen. Distorted and difficult to grasp each category. Therefore, by maintaining the solid angle and assigning it within a rectangular plane, it is possible to perform image operations such as image pasting work while equally looking at the omnidirectional image.
- FIG. 40 shows a rectangular display screen TM3 overlooking the projected image of the spherical screen. In this case, it is based on the development of a regular tetrahedron.
- Use screen ⁇ 4 that displays an area larger than 4 ⁇ so that the joint at the end of screen TM3 can be seen.
- screen ⁇ 3 is divided into six rectangles divided by line segment 165. The imaging area is displayed evenly. The shaded area 163 is one of them.
- the projection areas are equally displayed in 12 pentagons divided by dotted line 166.
- the shaded area 164 is one of them.
- the hatched portions 167, 168, and 169 are segmented regions of the input / output image in which the optical axis is set based on the regular icosahedron, regular tetrahedron, and regular octahedron, respectively. In this way, discontinuous seams in the input and output images can be found as desired, and differences in the exposure of each imaging area and projection area, temporal errors in the case of moving images, etc. can be corrected, and omnidirectional images can be accurately displayed on a spherical screen. Can be provided. Apply the above image manipulation and display method to the mapping method according to another embodiment.
- the display position of the spherical screen is a component force, an elevation angle, an azimuth angle, etc. You may show the scale. Use it for the 3D stereoscopic images currently being explored.
- Imaging is one of the mappings handled by the present invention.
- Figure 31 shows an illustration of an imaging method based on a cube.
- the large arc 170 divided into a solid angle region of 2/3 ⁇ sr indicates the edge of the spherical cube S7.
- the imager 60 is installed so that the optical center is at the center 05 of the spherical cube S7.
- Optical axis AX1 05 To the center 015 of PG1, one of the cubic square faces that touch S7.
- each image pickup device is set to 109.4712206 degrees or more so that the solid angle region of 2 / 3 ⁇ sr divided by the large arc 170 falls.
- the image picked up in this way is captured in a spherical cube, and then flattened into a rectangular plane based on the operation shown in the second embodiment.
- a line connecting the points 83, 014, 84, V8b, 81a on S7 with a large arc indicates a part of the above-mentioned equal product grid G7.
- a part of the grid G8 obtained by optically projecting the grid G7 from the optical center 05 is shown.
- An image captured by the image sensor 60 with G8 attached to and removed from PG1 as a vertical angle correction filter may be mapped to G6 in FIG.
- mapping handled by the present invention when the surface to be mapped is separated, for example, the image reflected on the mirror surface is reflected and mapped to the light receiving surface of the imaging device facing the surface at a distance. Including cases. In particular, an embodiment will be described in which the operation for integrating the images into the omnidirectional image is facilitated by avoiding the occurrence of a shift in the imaging region where the solid angle is divided.
- FIG. 32 is an explanatory diagram showing a regular tetrahedron as an example of a simultaneous imaging method in which optical centers are matched.
- the image pickup device 54 captures an image reflected on the reflector P12 with the optical axis 56 facing the reflector P12.
- the imager 54 and the reflector P12 are arranged so as to capture all directions while sharing the virtual optical center 012 with the other three imagers 54.
- Reflector P12 adjusts the curvature according to the angle of view of image sensor 54 and the distance from the virtual optical center. As shown in the drawing, if the imaging device 56 is arranged based on a regular tetrahedron, the reflector P12 is preferably a solid corresponding to this.
- FIG. 33 is a cross-sectional view including the optical axis 56.
- the surface REF1 constituting the reflector P12 forms a regular tetrahedron when the image pickup device 54-1 having an angle of view of about 141 ° or more in the sectional view 143 is used.
- the image pickup device 54-2 in the sectional view 144 is a case where a general-purpose camera having a general angle of view is used.
- Image incident on the surface REF2 constituting the anti hurts P12 is attached so that the curvature to provide an imaging area of 1 [pi sr to the imaging device ⁇ 4_ 2.
- the image pickup device 55 with the optical axis 56 facing outward may be paired with the image pickup device 54 and replaced with an image obtained by the image pickup device 55 having an angle of view that complements a portion where the image pickup device 54 is reflected.
- the present invention includes an imaging method that quickly obtains a rectangular image by multi-hierarchy mapping via a reflector while including multi-hierarchical mapping. Complex image processing after imaging can be reduced and moving images can be displayed in real time.
- Fig. 52 shows an image pickup device PG18 having two combinations of a reflector REF3 having a square pyramid base F18 and an image pickup device 225 arranged opposite to the reflector REF3.
- the two imagers 225 are set to capture the entire visual field imaging region in half via the reflector REF3, and share the virtual optical center 021. In this way, an omnidirectional square image can be created simply by pasting together two square images obtained by the imager PG18. There is no need to make corrections to share the two essences.
- a desired image may be acquired by re-mapping the captured image into a regular octahedron.
- the bottom F18 does not need to be square, but it is good.
- the rectangle has an aspect ratio of 3: 2 or 9: 8
- the back image is joined to the front image to obtain a rectangular image suitable for a general-purpose monitor with an aspect ratio of 3: 4 or 9:16.
- the aspect ratio is 2: 1, a square image can be obtained.
- the shape of the reflector in the present invention is not limited to a quadrangular pyramid, and may be applied to any solid.
- the arbitrary solid mentioned here includes a polyhedron including a regular polyhedron and a quasi-regular polyhedron, a solid including a hyperboloid, a solid including a curved surface, and a rotating body. For example, if the captured image has a reflector shape with a curved surface that can correct the solid angle, it is good.
- FIG. 49 is a top view of the cube imager PG6.
- a lens 216 is disposed along the optical axis, and each 2 ⁇ sr region 215 is imaged.
- the imaging regions 217 overlapped by the adjacent lenses 216 can be stereoscopically viewed as a parallax due to the optical center shift 181 of the two adjacent lenses 216.
- the lens used in the description of the sixteenth embodiment is a pinhole. May be substituted. It is also possible to share the light-receiving surface of multiple lenses, including a pair of lenses responsible for binocular parallax. In other words, it may be mapped alternately from a plurality of optical centers and optical axes on one mapping surface.
- One light receiving surface includes a three-dimensional light receiving surface composed of a plurality of surfaces. For example, if mapping to an ellipse, the diffuse reflection on the light-receiving surface is suppressed when setting the optical center at two focal points. Is preferred.
- the mapping handled by the present invention includes imaging, and the light receiving surface of the imaging device is regarded as a mapping surface at the time of input.
- An example in which a polyhedron is configured by a plurality of light receiving surfaces will be described.
- An omnidirectional image can be converted into a rectangular image by image processing, but the image quality is limited by image degradation and processing capability. This process can be simplified by making the light-receiving surface polyhedron and directly capturing the desired polyhedron image.
- FIG. 42 shows a regular tetrahedral imaging device PG9.
- a lens in this case a pinhole 189, is placed on each face of the regular tetrahedron.
- a triangle parallel to each surface of the regular tetrahedron is defined as a light receiving surface.
- the light receiving surface F8 is one of them.
- a regular tetrahedron image can be obtained without image processing by dispersing the optical axes in four directions and imaging on a regular tetrahedron-shaped light-receiving surface.
- the projection is geometrically pure, so that there is no distortion peculiar to the lens and all directions can be projected onto the polyhedral light receiving surface.
- FIG. 42 shows an explanatory diagram of the PG10 image sensor with a tetrahedron outline. Since the tetrahedron is cut off at the vertices and ridgelines, the tetrahedron is obtained, so that the light receiving surface F11 arranged according to the regular tetrahedron can be made into a hexagon, and the contact with the image circle C3 is improved.
- I / O terminals, power supplies 193, tripods and other attachments 194, and memory cards 195 can be attached to and detached from the surfaces F10 and F9 where ridgelines and vertices are notched.
- the light receiving element and display element such as CCD and CMOS are very fine when inputting and outputting information.
- it is handled as a divided mapping surface. Examples showing these shapes and arrangements are shown.
- the light receiving element along the subdivided equal product grid will be described. Since conventional light receiving elements are arranged in rectangular coordinates in rectangular coordinates, it is difficult to arrange RGB color filters evenly due to poor alignment with vertical transfer paths. Therefore, it is possible to arrange the light receiving elements into a polygon such as a hexagon, and to arrange them as evenly as possible by using a Z or equal product grid.
- FIG. 43 shows the arrangement rule 203 of the light receiving elements by enlarging the light receiving surface area.
- the light receiving elements 205 are arranged in accordance with the divided equal product grid G13.
- the vertical transfer path can be easily placed in the gap area 206 between the light receiving elements.
- the confusion circle of the microphone lens is easier to confine than the rectangular one, making it easier to prevent the CCD output from falling.
- the reference numerals “R”, “G”, and “B” written on the light receiving element 205 are shown to be evenly arranged by the force S indicating the arrangement of the three color filters and the arrangement of the three-way grid. .
- hexagonal light-receiving elements and three-way grids can be arranged without waste up to the edge of the light-receiving surface of a solid I in a three-dimensional surface area having a regular tetrahedron or other triangular surface than rectangular light-receiving elements or orthogonal grids.
- Arrangement rule 204 is an example of arrangement rule different from arrangement rule 203 by rotating light receiving elements 205 in the arrangement.
- the light receiving element may be circular like 205a. In the case of a circular light receiving element, it is easy to arrange a vertical transfer path in the gap region 206. Of course, other polygons may be arranged based on the grid. For example, when the regular dodecahedron is the light receiving surface, a pentagonal light receiving surface and a Z or pentagonal grid are preferable.
- the light receiving element may be replaced with a sensor that senses other electromagnetic waves, sound waves, temperature, the Doppler effect, or the like. Also, arrange using a grid different from the grid used for solid angle correction. Further, it may be replaced with a display element at the time of image output. The arrangement of the light receiving element and the display element may correspond as much as possible.
- a three-dimensional light receiving surface is arranged on one optical axis.
- the light receiving surface has been suitable to have a flat shape for filming, but in the case of a wide angle, the difference in exposure and distortion between the center and the periphery of the light receiving surface increases, and many lenses consist of multiple lenses. Therefore, we will explain how to use a cube as an example to provide a three-dimensional light receiving surface composed of multiple surfaces on one optical axis, and to make the distance between the light receiving surface and the focal point as equal as possible and to unify exposure as much as possible.
- FIG. 41 is an explanatory diagram of the image pickup device PG6.
- AX4 is one of the optical axes arranged on each surface. AX4 Place a pinhole at the upper point, for example, the intersection 178 of the cube surface. In this way, the omnidirectional image area is divided into six.
- the four triangles connecting the center 016 of the cube and the midpoint 172 between the vertices are designated as light receiving surface F6.
- the hatch part is one of the light receiving surfaces.
- the four triangles F7 connecting the center 016 of the cube and the vertex V16 are the light receiving surfaces, an angle of view close to 180 degrees can be obtained with one optical axis, depending on the pinhole accuracy. In other words, a wider angle image can be captured. In this way, a stereoscopic image can be created by overlapping a plurality of omnidirectional images and angles of view.
- the light receiving surface F6 may be replaced with a quadrangle having 172 as a vertex.
- Reference numeral 179 shows a cross-sectional view of the cubic imager.
- the three-dimensional surface of the light-receiving surface decreases as compared to the case where the light-receiving surface is a plane F6a with respect to the difference in the focal point distance to one point on the light-receiving surface.
- the obtained images are integrated according to the shape of the light receiving surface.
- PG7 shown in Fig. 57 is an omnidirectional image of a 24-hedron that is integrated according to the shape of the light-receiving surface F6.
- the quadrangular pyramid receiving surface F6 is mapped to a quadrangular pyramid F6b with 188 as the vertex and 184 as the bottom vertex.
- a rhombus dodecahedron image with geometric commonality can be mapped. After that, it is re-mapped into a cube, regular tetrahedron, regular octahedron, etc., and a rectangular planar image can be created by the same operation as in the other embodiments.
- a plurality of types of polyhedron images can be acquired by one imaging format, and an image optimized for various purposes can be provided.
- the present invention can be applied to other embodiments.
- the twentieth embodiment has been described in terms of geometrical expressions, it is within the range where the same effects as the contents of the above embodiments can be obtained.
- the theoretical values for explanation should be within the allowable range. Errors such as optical misalignment due to the size of the image pickup device, etc. may be approximated within the tolerance.
- the present invention includes a method of mapping an omnidirectional image onto a rectangular plane, but converting the light receiving surface, which is a mapping plane at the time of input, into a rectangular plane.
- Making an image pickup device that disperses and stereoscopically corresponds to a plurality of light receiving surfaces shown in the above-described embodiments is complicated. However, for example, using a regular octahedron, all light receiving surfaces can be integrated into a rectangular plane.
- Fig. 44 shows a conceptual diagram of a regular octahedron imager PG11. Pinholes or equivalent labels on each side 207 is installed.
- the partial area of the inner surface of PG25 is the light receiving surface, and an omnidirectional image can be taken.
- the light receiving surfaces installed on each face of the octahedron and formed into a triangular pyramid are integrated into three squares F13 sharing each vertex V25 of the regular octahedron. In this way, it is possible to provide a three-dimensionally arranged light receiving surface on each optical axis while integrating the light receiving surfaces into three rectangular planes.
- the concept of the light receiving surface may be replaced with a solid such as another polyhedron.
- the triangular pyramid surface region F12 is used as the light receiving surface so that the lens 207 can capture an image field of view of 1/2 ⁇ sr, one omnidirectional image can be obtained.
- the triangular pyramid surface area F12a with the imaging field of view extended to 1 ⁇ sr eight optical axes are distributed and two regular tetrahedral images are obtained.
- the entire surface of the square F13 is the light-receiving surface and the imaging field of view is expanded to 2 ⁇ sr (all-around fisheye)
- four omnidirectional images are obtained. If the adjacent optical axis overlaps the imaging area, this can be used to create a stereoscopic image.
- the surface F13a including F13 when used as a reflector, the three surfaces F13a are orthogonal to each other, so light coming from any direction can be returned to the light source (hereinafter referred to as an omnidirectional reflector).
- This principle can be used to quickly measure the position of imagers. However, when taking an image of 1/2 ⁇ sr or more with one optical axis, reflection occurs.
- Surface F13a may be used as a tripod to stabilize the imager.
- the light receiving surface may be arranged in a plane parallel to the surface of the octahedron instead of the triangular pyramid so that a regular octahedron image can be obtained.
- the present invention includes a method of using a plurality of imagers and covering the entire specific space. Observation with one omnidirectional imager is not possible due to the shape of the space in a curved space, and multiple imagers are required. An embodiment in which the positional relationship between the image pickup devices can be quickly grasped at this time will be described.
- FIG. 45 shows a conceptual diagram of a quasi-decahedron imaging device PG12.
- An optical axis is arranged based on each of the fourteen square faces F2 5 and images are taken by dividing all directions into six imaging areas.
- a point light source and a triangular pyramid reflector 208 are attached to the equilateral triangular surface F24 so as to be an omnidirectional reflector.
- Each surface of the reflector 208 intersects with each other at an angle of 90 degrees.
- the omnidirectional imager When multiple imagers with omnidirectional reflectors and light sources are scattered, the light of their own light source reflected by the other imagers is captured by the omnidirectional imager. The mutual positional relationship can be grasped. In this way, it is possible to analyze the omnidirectional images at each point and to observe the subject in the space to be observed from various angles.
- the allocation of the optical axis, the omnidirectional reflector, and the light source to the surface of the polyhedron is not limited to this.
- an omnidirectional image of a quasi-regular tetrahedron can be obtained.
- a quasi-regular tetrahedron image can be mapped on the front and back of the square 209 passing through the vertex V26 of the quasi-regular tetrahedron.
- a quasi-regular tetrahedron image can be mapped on the front and back of the hexagon 210 passing through the vertex V26.
- Planarized omnidirectional images can be arranged in a plane. However, when ensuring the continuity of the subject, there may be a planar array image with a gap. On the other hand, if the surface is filled, the subject may be partially discontinuous.
- An embodiment will be described in which the optical axis setting and the like are diversified and can be used for various purposes such as resolution and the presence or absence of stereoscopic imaging.
- An image pickup device PG14 whose conceptual diagram is shown in FIG. 47 has an image pickup module Ml having a lens 219 attached to a cube connected via a hinge 249. Therefore, module M 1 can rotate around AX6. Hinge 249 should be able to fix the rotation at a certain angle.
- a shirt 220, a battery and an adapter terminal 221, a display device 218 such as a monitor, and a connection device 248 are connected to one imaging device PG14.
- the connection device 248 is a good one that also serves as a signal and power input / output and a fixture for a tripod.
- the lens 219 may be replaced with a pinhole.
- the light receiving surface 222 is a part of a spherical surface. If this is a three-dimensional light receiving surface, It is difficult to control the diffuse reflection of light at the center, especially when using a pinhole and the light-receiving surface is made spherical, so the incident light 251 hits the light-receiving surface perpendicularly and is suitable for light reception and in the opposite direction of the incident light. Since the light is reflected, the reflected light 252 need not be irregularly reflected.
- the internal space 223 contains the elements necessary for the imager.
- Fig. 48 shows an example 253 in which two imaging modules Ml are aligned and used as a stereoscopic image pickup device, and an example 254 in which the optical axes are turned in opposite directions and used as an omnidirectional image pickup device. .
- Fig. 50 shows a usage example in which three units are combined.
- Imager PG14 is connected to other imagers in a state where module Ml is rotated 90 times and connected.
- the optical axis arrangement is based on the cube PG15, and the module Ml picks up an image of the entire field of view divided into six.
- three PG14s can capture omnidirectional images with higher image quality than a single imager.
- the element module M2 which is necessary for imaging, such as a light source and a pan head, can be attached to and removed from the recessed part of the cube PG15. In addition, it is good if it is easy to install by cutting out the top 224.
- module M 1 may be set freely.
- the connecting device 248 may be increased or replaced with a hinge.
- Module Ml always needs to be paired.
- Various combinations of imaging elements can be used for various purposes.
- the optical axis setting can be varied.
- an embodiment will be described in which the polyhedron imager is deformed without impairing the geometrical characteristics of the polyhedron on which the optical axis is set, and the necessary components such as a memory and an electronic circuit are incorporated in the imager.
- FIG. 46 is a conceptual diagram of the icosahedron PG13 in which the ridgeline and the apex of the octahedron 211 are cut out and a cross-sectional view F14a thereof.
- the 26-sided imaging device PG13 has optical axes arranged on at least eight triangular surfaces with the light receiving surfaces concentrated on the six octagonal surfaces F14 passing through the vertex V27.
- the hatched portion F15 of the octagonal surface F14 becomes a part of the imaging surface of the optical axis AX15.
- an omnidirectional image equivalent to the octahedral imager that is the basis of the optical axis arrangement is obtained.
- the elements necessary for the imaging device such as ram, memory, and power supply are stored in the internal space 212 separated by the octagonal plane F14.
- internal space A detachable device 214 such as an image pickup device can be attached to 213.
- Various omnidirectional images can be obtained by distributing the optical axis using a 26-hedron. For example, if the optical axes are arranged on all surfaces, a 26-hedron image can be obtained and each angle of view ⁇ 5 can be handled by a general-purpose lens.
- An omnidirectional image of a rhomboid dodecahedron can be obtained by arranging optical axes on 12 rectangular planes F17 formed by cutting out the ridgeline of the octahedron 211.
- optical axes are arranged on the six rectangular surfaces F16 that are created by cutting out the vertices of the octahedron 211, and the region F15 is replaced with a reflector to make an omnidirectional reflector, and the remaining surface has a light source, a tripod, and an output. Detachable devices such as input plugs and antennas may be incorporated. In addition, by arranging the optical axes on eight surfaces aligned in the orthogonal direction of the axis AX6, three octagonal or cylindrical images can be obtained and reduced to a general panoramic image.
- mappings handled by the present invention include those in which different mapping methods are mixed, such as a vertical projection in the vertical direction and a parallel projection in the horizontal direction.
- the target solids include solids with open surfaces and curved surfaces, and combinations of multiple solids.
- cylindrical projections are combined to extract regions with relatively little distortion and integrate them into a solid with one surface closed.
- FIG. 53 is a conceptual diagram showing a process of obtaining a square image by intersecting two cylinders PG19.
- An omnidirectional image by parallel projection and cardiac projection is pasted on the cylindrical PG19 with the left and right sides open.
- F19 is the overlapping area that intersects another cylinder PG19.
- the hatched part F19 is extracted from the cylindrical image and used.
- the other cylindrical region PG20a is removed, and a three-dimensional image PG20 is obtained by the region F19.
- the solid PG20 has the graticule of the earth for explanation.
- the three-dimensional image PG20 is mapped onto the front and back surfaces of the square F20 from the vertical direction along the axis AX8. Rotate the image mapped on the back side and integrate it with the front side to get a rectangular omnidirectional image
- the curved surface of the solid PG20 by this mapping method is composed of a part of a cylinder, there is an advantage that the entire field of view can be squared with little distortion while being similar to a general panorama image.
- the solid defined by the curved surface, straight line, and curved line is not limited to the description of the twenty-fifth embodiment. For example, when a plurality of cones are combined, an image similar to the Lambert equirectangular projection can be obtained.
- Fig. 54 shows a rhombus dodecahedron PG23 composed of curved surfaces. This is obtained by adding a cylinder to PG20 shown in Fig. 53 and intersecting three cylinders. Therefore, the surface of the solid PG23 is composed of a cylinder. The 12 curved surfaces divided by the arc 238 connecting the vertices V29 are congruent. If this solid is divided by the straight line 239 connecting the vertices V29, a 24-hedron is obtained.
- This three-dimensional PG23 can be easily mapped onto the polyhedron with many parts sharing the arrangement of vertices, ridges, faces, and the like with the rhomboid dodecahedron, cube, tetrahedron, and octahedron.
- the hatched area 240 is mapped to one face of a regular octahedron.
- “easy” includes the advantage that the polyhedron can be obtained only by projecting in an epicenter and each surface can be corrected for solid angles. If these polyhedrons are used, a rectangular plane can be obtained by the planarization method of the present invention. In this way, a general-purpose 360-degree cylindrical image can be compromised.
- the present invention is not limited to the twenty-sixth embodiment, but can be applied to other embodiments. However, although the above-described twenty-sixth embodiment has been described in terms of geometrical expressions, it can be applied as long as the same effects as the contents of the above-described embodiments can be obtained.
- the “regular polyhedron” used in the description of the twenty-sixth embodiment may be replaced with “polyhedron”, and “regular polygon” may be replaced with “polygon 'shape”.
- the meridian of the sphere is an arc. Therefore, it is possible to obtain a rectangular plane on which a subject can be smoothly displayed via a solid whose specific cross section is a rectangle and all cross sections in the meridian direction are arcs.
- the solid PG21 shown in the conceptual diagram of Fig. 58 has its surface divided by the graticule grid G15.
- the graticule grid G15 consists of meridians 226 and 231.
- the grid G15 shown in the figure displays an image area of l / 2 sr, for example, only 1/4 of the northern hemisphere.
- the arc 226 passes through the pole 227 and the point 230 on the edge 228 of the square F21, and the curve 231 connects the points dividing the arc 226.
- grid G16 is on square F21a, and is composed of meridian 226a, parallel 231a and square ridge 228a corresponding to the equator.
- the grid G16 shown in the figure shows only a hemisphere, that is, a solid angle region of 2 ⁇ sr.
- the line segment 226a is a line segment passing through the pole 227a on the square and the point 230a on the square edge 228a.
- Line segment 231a is a chained curve that connects the points dividing line segment 226a. is there. In this way, the divided region 233 of the grid G15 is mapped to the corresponding region 233a on the grid G16 to obtain a square image.
- Three-dimensional PG21 used for mapping is a shape in which latitudes are chained together by curves and adjacent surfaces are smoothly continuous. Therefore, the subject will not be bent unnaturally.
- the present invention includes segmentation by curves including quadratic curves.
- the square grid G17 whose conceptual diagram is shown in Fig. 59, shows the entire field of view, that is, the solid angle region of 4 ⁇ sr.
- the grid G16 for 2 ⁇ sr shown in Fig. 58 corresponds to the square area F21b.
- Grid G17 is composed of line segment 226b, line segment 231b and square edge 228b.
- Line segment 226b is a linked hyperbola passing through pole 227b and point 230b on the square edge 228b.
- Line segment 231b is a linked parabola connecting the points that segment line 226b.
- the curvature and the like are adjusted so that the line segment 226b and the line segment 231b are orthogonal to each other at the intersection point.
- the segmented area 233 on the grid G15 is mapped to the corresponding area 233b of the grid G17.
- Connecting the meridian lines with a curve makes the subject look natural without bending over the ridgeline 228b.
- Grids 15, 16, and 17 are adjusted so that image area 233 is an equal product map.
- Dividing lines are geodesic lines, arcs, Beziers, spline curves, 4 blue arcs, quadratic curves such as parabola, cubic curves such as elliptical curves, quaternary curves such as Cassinian oval, relaxation curves such as clothoid curves. including. These are mixed and chained.
- the present invention includes a product obtained by equally dividing an equal product division into a spherical product or a polyhedron and a curved surface that divides a rectangular plane.
- Figure 55 shows a regular octahedron image with the subject displayed smoothly. It is a conceptual diagram which shows the process in which multi-layer mapping to a square is considered in consideration.
- the regular octahedron image PG22 is mapped to a solid PG22a composed of curved surfaces, and then mapped to a square image F22.
- the regular octahedron image region PG 22 is divided into 24 equal parts by grid G18. Region 235 is one of them.
- Grid G18 is composed of line segment 236 connecting vertex V28 and midpoint 234 of the ridgeline, and ridgeline 237.
- the solid PG22a shares the vertex V28 and the center 022 except for the vertex V28a with the regular octahedron PG22.
- Each face is divided into 24 equal parts by grid G18a, and region 235a is one of them.
- Grid G18a has a line segment connecting vertices V28a and V28, a line segment connecting 237a and V28's midpoint 234 and vertex V28a, and a line connecting two vertices V28 and a point on line 237a 236a Consists of.
- the line segment 236a is a curved line such as a chained arc.
- the curvature of the line segment 236a should be adjusted so that the contour force of the square does not protrude when mapping to a square with V28 as the vertex.
- Square F22 shares vertex V28 and center 022 with octahedron PG22.
- the square is divided into 24 equal parts by grid G18b, and region 235b is one of them.
- Grid G18b is composed of line segment 237b connecting point 022 and vertex V28, line segment connecting point 022 and midpoint 234 of the edge, and line segment 236b connecting two vertices V28 and a point on line segment 237b.
- the regular octahedron image PG22 is mapped to the square F22 via the solid PG22a. If an equivalent effect can be obtained, mapping from a regular octahedron to a square may be performed.
- FIG. 56 is a conceptual diagram of a cylindrical image pickup device PG24.
- the imaging unit 241 makes a pair in the direction of the axis AX11 of the cylinder.
- the imaging unit 241 includes a lens 242 and a reflector 244.
- Around the lens 242 is disposed a reflector 244 having a rotating body shape with an axis AX11. Reflecting the 360-degree horizontal and vertical ⁇ 8 fields of view from the virtual optical center 246, the lens 242 is sent. Therefore, the lens 242 only needs to have an angle of view ⁇ 6 that can capture an image reflected on the reflector 244.
- the two imaging units 241 maintain an appropriate distance L5 by means of a transparent cylinder 245, etc.
- the distance L5 depends on the angle of view of the lens 242 and the angle ⁇ 9 of the reflector 244. Therefore, it is possible to zoom by sliding the cylinder 245 or to use the small PG24a at the time of insertion.
- the omnidirectional images taken by the two imagers in this way can be made rectangular by the planarization method of the present invention.
- the two optical centers 246 of the paired imaging unit 241 provide a stereoscopic image and can be observed stereoscopically.
- the imaging unit can be incorporated in the rotating body of the other reflector, space can be saved. When used in a narrow space, there are advantages such that the distance required for image formation can be obtained via a reflector, and the lens is too close to the inner wall of the space to prevent imaging.
- the shaded area 247 has a built-in memory and transmission / reception system, or an imaging unit 241a is added, and if the angle of view ⁇ 60 is reflected in the lens 242 and an appropriate shooting angle cannot be obtained, the imaging unit 242 is desired.
- the imaging of the subject to be performed can be substituted.
- the four imaging units are arranged side by side in the direction of the force S and the axis AX11 incorporated into the imaging device PG24, the diameter D1 of the cylinder does not increase, and the minimum diameter of the observation tubular space is limited.
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008552197A JP4945578B2 (ja) | 2007-01-04 | 2007-12-25 | 情報処理方法 |
CN2007800493257A CN101606177B (zh) | 2007-01-04 | 2007-12-25 | 信息处理方法 |
US12/497,727 US8665273B2 (en) | 2007-01-04 | 2009-07-06 | Method of mapping image information from one face onto another continuous face of different geometry |
US14/158,880 US9519995B2 (en) | 2007-01-04 | 2014-01-20 | Method of mapping image information from one face onto another continuous face of different geometry |
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2007-023528 | 2007-01-04 | ||
JP2007023528 | 2007-01-04 | ||
JP2007106534 | 2007-03-16 | ||
JP2007-106534 | 2007-03-16 | ||
JP2007271821 | 2007-09-19 | ||
JP2007-271821 | 2007-09-19 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/497,727 Continuation-In-Part US8665273B2 (en) | 2007-01-04 | 2009-07-06 | Method of mapping image information from one face onto another continuous face of different geometry |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2008081961A1 true WO2008081961A1 (ja) | 2008-07-10 |
Family
ID=39588636
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2007/075356 WO2008081961A1 (ja) | 2007-01-04 | 2007-12-25 | 情報処理方法 |
Country Status (4)
Country | Link |
---|---|
US (2) | US8665273B2 (ja) |
JP (1) | JP4945578B2 (ja) |
CN (1) | CN101606177B (ja) |
WO (1) | WO2008081961A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014137891A (ja) * | 2013-01-16 | 2014-07-28 | Hitachi Vehicle Energy Ltd | 角形二次電池 |
WO2017029885A1 (ja) * | 2015-08-18 | 2017-02-23 | 株式会社ソニー・インタラクティブエンタテインメント | 画像生成装置、及び画像表示制御装置 |
Families Citing this family (65)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4945578B2 (ja) | 2007-01-04 | 2012-06-06 | 肇 鳴川 | 情報処理方法 |
TW200949211A (en) * | 2008-06-04 | 2009-12-01 | Tomtom Int Bv | Method and apparatus for preparing map data |
US8243075B2 (en) * | 2008-10-14 | 2012-08-14 | Autodesk, Inc. | Graphics processing unit accelerated dynamic radial tessellation |
US20110175905A1 (en) * | 2010-01-15 | 2011-07-21 | American Propertunity, LLC. | Infoshape: displaying multidimensional information |
EP2388987A1 (en) * | 2010-05-19 | 2011-11-23 | Thomson Licensing | Camera with volumetric sensor chip |
US8896601B1 (en) * | 2010-10-12 | 2014-11-25 | Google Inc. | Projecting geographic data from a spherical surface to two-dimensional cartesian space |
US9842334B1 (en) * | 2011-12-30 | 2017-12-12 | EMC IP Holding Company LLC | Identifying risky transactions |
US8902322B2 (en) * | 2012-11-09 | 2014-12-02 | Bubl Technology Inc. | Systems and methods for generating spherical images |
US10165157B2 (en) * | 2013-02-19 | 2018-12-25 | Disney Enterprises, Inc. | Method and device for hybrid robotic/virtual pan-tilt-zoom cameras for autonomous event recording |
US8699868B1 (en) * | 2013-03-14 | 2014-04-15 | Microsoft Corporation | Anti-shake correction system for curved optical sensor |
US10365804B1 (en) * | 2014-02-20 | 2019-07-30 | Google Llc | Manipulation of maps as documents |
US10078898B2 (en) * | 2014-11-07 | 2018-09-18 | National Institute Of Standards And Technology | Noncontact metrology probe, process for making and using same |
US9536328B2 (en) | 2015-01-06 | 2017-01-03 | International Business Machines Corporation | Method to optimize the visualization of a map's projection based on data and tasks |
US9665989B1 (en) * | 2015-02-17 | 2017-05-30 | Google Inc. | Feature agnostic geometric alignment |
EP3338176A1 (en) * | 2015-08-17 | 2018-06-27 | Ricoh Company, Ltd. | Wide view image display system, information processing apparatus, and image display method |
JP2017040687A (ja) | 2015-08-17 | 2017-02-23 | 株式会社リコー | 画像表示システム、情報処理装置、画像表示方法及びプログラム |
US9995848B2 (en) | 2015-09-16 | 2018-06-12 | International Business Machines Corporation | Adaptive placement of weather sensors in response to dynamic local conditions |
US9858706B2 (en) | 2015-09-22 | 2018-01-02 | Facebook, Inc. | Systems and methods for content streaming |
US10096130B2 (en) | 2015-09-22 | 2018-10-09 | Facebook, Inc. | Systems and methods for content streaming |
TW201719572A (zh) * | 2015-11-19 | 2017-06-01 | 國立交通大學 | 三維模型分析及搜尋方法 |
US10282814B2 (en) * | 2016-01-07 | 2019-05-07 | Mediatek Inc. | Method and apparatus of image formation and compression of cubic images for 360 degree panorama display |
WO2017127816A1 (en) * | 2016-01-22 | 2017-07-27 | Ziyu Wen | Omnidirectional video encoding and streaming |
CN107040779B (zh) * | 2016-02-03 | 2020-04-21 | 腾讯科技(深圳)有限公司 | 全景视频编码方法及装置 |
US10157448B2 (en) * | 2016-02-12 | 2018-12-18 | Qualcomm Incorporated | Foveated video rendering |
US10319071B2 (en) | 2016-03-23 | 2019-06-11 | Qualcomm Incorporated | Truncated square pyramid geometry and frame packing structure for representing virtual reality video content |
WO2017175888A1 (ko) * | 2016-04-05 | 2017-10-12 | 삼성전자 주식회사 | 이미지 처리 방법 및 장치 |
US10249019B2 (en) * | 2016-05-06 | 2019-04-02 | Mediatek Inc. | Method and apparatus for mapping omnidirectional image to a layout output format |
US10848668B2 (en) | 2016-05-19 | 2020-11-24 | Avago Technologies International Sales Pte. Limited | 360 degree video recording and playback with object tracking |
US10681326B2 (en) * | 2016-05-19 | 2020-06-09 | AVAGO TECHNOLOGlES INTERNATIONAL SALES PTE. LIMITED | 360 degree video system with coordinate compression |
US11019257B2 (en) | 2016-05-19 | 2021-05-25 | Avago Technologies International Sales Pte. Limited | 360 degree video capture and playback |
KR102506480B1 (ko) * | 2016-06-14 | 2023-03-07 | 삼성전자주식회사 | 영상 처리 장치 및 그 영상 처리 방법 |
US9971556B2 (en) * | 2016-07-05 | 2018-05-15 | Ricoh Company, Ltd. | Image processing apparatus, image processing method, and recording medium |
CN109478347A (zh) * | 2016-07-29 | 2019-03-15 | 索尼公司 | 图像处理装置和图像处理方法 |
CN106341673A (zh) * | 2016-08-15 | 2017-01-18 | 李文松 | 一种新型2d/3d全景vr视频的存储方法 |
JP6657475B2 (ja) * | 2016-08-25 | 2020-03-04 | エルジー エレクトロニクス インコーポレイティド | 全方位ビデオを伝送する方法、全方位ビデオを受信する方法、全方位ビデオの伝送装置及び全方位ビデオの受信装置 |
WO2018043905A1 (ko) * | 2016-08-29 | 2018-03-08 | 엘지전자 주식회사 | 360 비디오를 전송하는 방법, 360 비디오를 수신하는 방법, 360 비디오 전송 장치, 360 비디오 수신 장치 |
US10127637B2 (en) * | 2016-08-30 | 2018-11-13 | Adobe Systems Incorporated | Automatic orientation adjustment of spherical panorama digital images |
US20180075576A1 (en) * | 2016-09-09 | 2018-03-15 | Mediatek Inc. | Packing projected omnidirectional videos |
US10462484B2 (en) | 2016-10-07 | 2019-10-29 | Mediatek Inc. | Video encoding method and apparatus with syntax element signaling of employed projection layout and associated video decoding method and apparatus |
WO2018064965A1 (en) | 2016-10-07 | 2018-04-12 | Mediatek Inc. | Method and apparatus for generating projection-based frame with 360-degree image content represented by triangular projection faces assembled in octahedron projection layout |
KR102371099B1 (ko) | 2016-11-17 | 2022-03-04 | 인텔 코포레이션 | 광시야 비디오를 인코딩하기 위한 구면 회전 기법 |
BR112019007279A2 (pt) * | 2016-11-17 | 2019-07-09 | Intel Corp | indicação de janela de visualização sugerida para vídeo panorâmico |
KR20180059210A (ko) * | 2016-11-25 | 2018-06-04 | 삼성전자주식회사 | 영상 처리 장치 및 그 영상 처리 방법 |
US10380715B2 (en) | 2016-12-07 | 2019-08-13 | Mediatek Inc. | Method and apparatus for generating and encoding projection-based frame with 360-degree content represented by triangular projection faces packed in octahedron projection layout |
CN110115026B (zh) * | 2016-12-19 | 2021-07-13 | 三星电子株式会社 | 在电子装置中在矩形投影上产生360度内容的方法和系统 |
WO2018131813A1 (en) * | 2017-01-10 | 2018-07-19 | Samsung Electronics Co., Ltd. | Method and apparatus for generating metadata for 3d images |
KR102503342B1 (ko) * | 2017-01-10 | 2023-02-28 | 삼성전자주식회사 | 스테레오 스코픽 비디오 콘텐츠를 전송하기 위한 방법 및 장치 |
WO2018131803A1 (ko) * | 2017-01-10 | 2018-07-19 | 삼성전자 주식회사 | 스테레오 스코픽 비디오 콘텐츠를 전송하기 위한 방법 및 장치 |
US10580167B1 (en) * | 2017-01-24 | 2020-03-03 | Amazon Technologies, Inc. | Excluding masked regions of virtual reality (VR) frames from encoder processing |
US10593122B1 (en) * | 2017-01-24 | 2020-03-17 | Amazon Technologies, Inc. | Mirroring edge pixels |
KR20190129865A (ko) * | 2017-03-24 | 2019-11-20 | 소니 주식회사 | 정보 처리 장치 및 정보 처리 방법, 그리고 프로그램 |
WO2018190446A1 (ko) * | 2017-04-13 | 2018-10-18 | 삼성전자 주식회사 | 영상 처리 방법 및 장치 |
KR102454161B1 (ko) * | 2017-06-15 | 2022-10-14 | 삼성전자주식회사 | 전방향 이미지를 프로세싱하기 위한 방법 및 장치 |
CN109559631A (zh) * | 2017-09-26 | 2019-04-02 | 李兴奇 | 圆形世界地图的制作方法 |
KR102442089B1 (ko) * | 2017-12-20 | 2022-09-13 | 삼성전자주식회사 | 이미지의 관심 영역의 위치에 따라 이미지를 다면체에 매핑하는 장치 및 그 영상 처리 방법 |
CN108366249A (zh) * | 2018-01-17 | 2018-08-03 | 安徽中骄智能科技有限公司 | 一种基于wifi模块的虚拟现实vr数据快速处理系统 |
CN111684793A (zh) * | 2018-02-08 | 2020-09-18 | 索尼公司 | 图像处理装置、图像处理方法、程序以及投影系统 |
US11012657B2 (en) * | 2018-06-08 | 2021-05-18 | Lg Electronics Inc. | Method for processing overlay in 360-degree video system and apparatus for the same |
US11074753B2 (en) * | 2019-06-02 | 2021-07-27 | Apple Inc. | Multi-pass object rendering using a three- dimensional geometric constraint |
GB2585645B (en) * | 2019-07-08 | 2024-04-17 | Toshiba Kk | Computer vision method and system |
CN110401040B (zh) * | 2019-07-26 | 2020-05-26 | 中国电子科技集团公司第五十四研究所 | 一种基于等化面积和混合形状的反射面天线曲面分块方法 |
CN111240625B (zh) * | 2020-01-09 | 2022-03-18 | 盾钰(上海)互联网科技有限公司 | 一种无限视觉边界的图像动态渲染的计算方法及系统 |
CN114549671A (zh) * | 2020-11-26 | 2022-05-27 | 华为技术有限公司 | 一种网格编码方法及计算机系统 |
RU205591U1 (ru) * | 2021-04-05 | 2021-07-22 | Денис Владимирович Гаврилов | Голографический октаэдр |
CN113256485B (zh) * | 2021-05-21 | 2024-01-30 | 百果园技术(新加坡)有限公司 | 图像拉伸方法、装置、电子设备和存储介质 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62502515A (ja) * | 1985-03-15 | 1987-10-01 | スピラウス ア−セルスト−ン | 周期的なモザイク構造を有する地図パズル |
JPH03205082A (ja) * | 1989-12-31 | 1991-09-06 | Toshiaki Betsumiya | 菱形多面体の創り方 |
JP2006309802A (ja) * | 2006-08-17 | 2006-11-09 | Sony Corp | 画像処理装置及び画像処理方法 |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US2393676A (en) * | 1944-02-25 | 1946-01-29 | Fuller Richard Buckminster | Cartography |
US6141034A (en) * | 1995-12-15 | 2000-10-31 | Immersive Media Co. | Immersive imaging method and apparatus |
JP2002083285A (ja) * | 2000-07-07 | 2002-03-22 | Matsushita Electric Ind Co Ltd | 画像合成装置および画像合成方法 |
JP3975736B2 (ja) | 2001-12-07 | 2007-09-12 | ソニー株式会社 | 画像処理装置及び画像処理方法、記憶媒体、並びにコンピュータ・プログラム |
CA2476610A1 (en) * | 2002-02-27 | 2003-09-04 | Geo-Rae Co., Ltd. | Method and system for controlling a stereoscopic camera |
JP2004272590A (ja) * | 2003-03-07 | 2004-09-30 | Sony Corp | データ符号化装置及びデータ符号化方法、並びにコンピュータ・プログラム |
CN1271572C (zh) * | 2004-08-05 | 2006-08-23 | 上海交通大学 | 二维图像序列三维重建方法 |
US7686616B1 (en) * | 2006-01-24 | 2010-03-30 | William Larry Cloud | Globe reversibly convertible to a map |
JP4945578B2 (ja) | 2007-01-04 | 2012-06-06 | 肇 鳴川 | 情報処理方法 |
US8638327B2 (en) * | 2007-11-14 | 2014-01-28 | Microsoft Corporation | Tiled projections for planar processing of round earth data |
-
2007
- 2007-12-25 JP JP2008552197A patent/JP4945578B2/ja active Active
- 2007-12-25 WO PCT/JP2007/075356 patent/WO2008081961A1/ja active Application Filing
- 2007-12-25 CN CN2007800493257A patent/CN101606177B/zh active Active
-
2009
- 2009-07-06 US US12/497,727 patent/US8665273B2/en not_active Expired - Fee Related
-
2014
- 2014-01-20 US US14/158,880 patent/US9519995B2/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62502515A (ja) * | 1985-03-15 | 1987-10-01 | スピラウス ア−セルスト−ン | 周期的なモザイク構造を有する地図パズル |
JPH03205082A (ja) * | 1989-12-31 | 1991-09-06 | Toshiaki Betsumiya | 菱形多面体の創り方 |
JP2006309802A (ja) * | 2006-08-17 | 2006-11-09 | Sony Corp | 画像処理装置及び画像処理方法 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2014137891A (ja) * | 2013-01-16 | 2014-07-28 | Hitachi Vehicle Energy Ltd | 角形二次電池 |
WO2017029885A1 (ja) * | 2015-08-18 | 2017-02-23 | 株式会社ソニー・インタラクティブエンタテインメント | 画像生成装置、及び画像表示制御装置 |
JPWO2017029885A1 (ja) * | 2015-08-18 | 2018-04-26 | 株式会社ソニー・インタラクティブエンタテインメント | 画像生成装置、及び画像表示制御装置 |
RU2686591C1 (ru) * | 2015-08-18 | 2019-04-29 | Сони Интерактив Энтертейнмент Инк. | Устройство выработки изображения и устройство управления отображением изображения |
US10659742B2 (en) | 2015-08-18 | 2020-05-19 | Sony Interactive Entertainment, Inc. | Image generating apparatus and image display control apparatus |
Also Published As
Publication number | Publication date |
---|---|
JPWO2008081961A1 (ja) | 2010-04-30 |
US20140132598A1 (en) | 2014-05-15 |
CN101606177B (zh) | 2013-07-17 |
US20100001997A1 (en) | 2010-01-07 |
US8665273B2 (en) | 2014-03-04 |
CN101606177A (zh) | 2009-12-16 |
US9519995B2 (en) | 2016-12-13 |
JP4945578B2 (ja) | 2012-06-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4945578B2 (ja) | 情報処理方法 | |
US7271803B2 (en) | Method and system for simulating stereographic vision | |
TWI356186B (en) | Panoramic video system with real-time distortion-f | |
US9609210B2 (en) | Technique for view synthesis | |
US6323862B1 (en) | Apparatus for generating and interactively viewing spherical image data and memory thereof | |
JP5054971B2 (ja) | ディジタル3d/360度カメラシステム | |
EP0522204B1 (en) | Method and apparatus for dodecahedral imaging system | |
KR100882011B1 (ko) | 회전 대칭형의 광각 렌즈를 이용하여 전방위 영상을 얻는 방법 및 장치 | |
CN105245838A (zh) | 一种全景视频播放方法及播放器 | |
JP2010033573A (ja) | 情報処理方法 | |
KR102072639B1 (ko) | 화상 생성 장치 및 화상 표시 제어 장치 | |
TW200425737A (en) | Method for creating brightness filter and virtual space creation system | |
EP2685707A1 (en) | System for spherical video shooting | |
JP2022116089A (ja) | 撮像システム、方法及び用途 | |
Gurrieri et al. | Acquisition of omnidirectional stereoscopic images and videos of dynamic scenes: a review | |
CN110211028B (zh) | 一种基于球面全景显示的鱼眼图像显示方法 | |
JP2007013514A (ja) | 720°全方位矩形画像モニタリングシステム | |
US7148891B2 (en) | Image display method and image display device | |
JP6682984B2 (ja) | 自由視点映像表示装置 | |
US20220253975A1 (en) | Panoramic presentation methods and apparatuses | |
KR101889225B1 (ko) | 입체 전방위 영상 획득 방법과 재생 방법 및 입체 전방위 카메라 | |
TWI457692B (zh) | 立體全景影像技術 | |
CN113066154B (zh) | 一种地表图像与地下空间图像实时叠加的方法和系统 | |
Guan | Spherical image processing for immersive visualisation and view generation | |
JP2020024687A (ja) | 情報処理装置、情報処理システム、情報処理方法及びプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200780049325.7 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 07860551 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008552197 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 07860551 Country of ref document: EP Kind code of ref document: A1 |