Collaborative disparity decomposition
A novel disparity computation technique is presented which comprises multiple orthogonal disparity maps, generated from approximately orthogonal decomposition feature spaces, collaboratively generating a composite disparity map. Using an approximately orthogonal feature set extracted from such feature spaces produces an approximately orthogonal set of disparity maps that can be composited together to produce a final disparity map. Various methods for dimensioning scenes and are presented. One approach extracts the top and bottom vertices of a cuboid, along with the set of lines, whose intersections define such points. It then defines a unique box from these two intersections as well as the associated lines. Orthographic projection is then attempted, to recenter the box perspective. This is followed by the extraction of the three-dimensional information that is associated with the box, and finally, the dimensions of the box are computed. The same concepts can apply to hallways, rooms, and any other object.
1. A method for defining a unique object, comprising the steps of:
imaging an object employing a plurality of images;
defining one or more of the object's vertices;
matching the one or more defined vertices across a plurality of the plurality of images to extract a three-dimensional vector for each of the one or more defined vertex combinations;
extracting a line segment connecting a first vertex of the one or more defined vertices and a second vertex of the object;
projecting a first vertex of the object and its associated edges to a canonical system;
projecting the extracted three-dimensional vector associated with the second vertex also into the canonical system;
extracting a z-component of one or more of the sides defining the first vertex of the object;
projecting the extracted line segment to the canonical system;
extracting one or more geometric relationships between one or more sides and the line segment of the object; and
dimensioning the object in accordance with at least the one or more extracted angular relationships and the extracted line segment.
2. The method of claim 1 , further comprising the step of receiving an input identifying one or more of the vertices of the object.
3. The method of claim 1 , further comprising the step of matching vanishing lines from one or more of the plurality of images to identify the first vertex.
4. The method of claim 3 , further comprising the step of matching vanishing lines from one or more of the plurality of images to define the second vertex.
5. The method of claim 1 , wherein the object is an irregularly shaped box.
6. The method of claim 1 , wherein the object is an irregularly shaped room.
7. The method of claim 1 , wherein the object is an irregularly shaped object.
8. The method of claim 1 , further comprising the steps of defining a depth map to identify the first vertex.
9. The method of claim 8 , further comprising the steps of:
defining a first component depth map having a first dimension;
defining a second component depth map having a second dimension, the second dimension being substantially orthogonal to the first dimension; and
combining the first and second depth maps to provide the depth map.
10. The method of claim 1 , wherein the plurality of images are right land left images of a stereo pair.
11. The method of claim 1 , wherein three or more images are employed to image the object.
12. The method of claim 1 , further comprising the step of modifying the dimensioned object in accordance with a global model.
13. The method of claim 1 , further comprising the steps of projecting a structured light pattern on the object; determining one or more discontinuities in the structured light pattern; and determining one or more edges of the object in accordance with the one or more determined discontinuities.
14. A system for defining a unique object, comprising:
an imaging apparatus for imaging an object; and
a processor for defining one or more of the vertices representative of the object; matching the one or more defined vertices across left and right viewpoints to extract a three-dimensional vector for each of the one or more defined vertices; extracting a line segment connecting a first vertex of the one or more defined vertices and a second vertex of the object; projecting a first vertex representative of the object and its associated edges to a canonical system; projecting the extracted three-dimensional vector associated with the first vertex also into the canonical system; extracting a z-component of one or more of the sides defining the first vertex of the object; projecting the extracted line segment to the canonical system; extracting one or more geometric relationships between one or more sides and the line segment of the object; and dimensioning the object in accordance with at least the one or more extracted angular relationships and the extracted line segment.
15. The system of claim 14 , wherein the imaging apparatus is a stereo imaging apparatus.
16. The system of claim 15 , wherein the imaging apparatus employs left and right images imaged by the stereo imaging apparatus.
17. The system of claim 14 , wherein the object is an irregular box.
18. The system of claim 14 , wherein the object is an irregular object.
19. The system of claim 14 , further comprising an input device for receiving an input indicating a location of one or more vertices of the object.
20. The system of claim 14 , wherein the processor further matches vanishing lines from one or more of the first and second images to identify one or more of the first and second first vertices.
21. The system of claim 14 , further comprising: a structured light projector projecting a structured light pattern on the object; and wherein the processor further determines one or more discontinuities in the structured light pattern, and further determines one or more edges of the object in accordance with the one or more determined discontinuities.
22. A non-transitory computer readable storage medium having a computer program stored thereon, the computer program causing a general purpose computer to perform the steps of:
defining one or more of the object's vertices;
matching the one or more defined vertices across left and right viewpoints to extract a three-dimensional vector for each of the one or more defined vertices;
extracting a line segment connecting a first vertex of the one or more defined vertices and a second vertex of the object;
projecting a first vertex of the object and its associated edges to a canonical system;
projecting the extracted three-dimensional vector associated with the first vertex also into the canonical system;
extracting a z-component of one or more of the sides defining the first vertex of the object;
projecting the extracted line segment to the canonical system;
extracting one or more geometric relationships between one or more sides and the line segment of the object; and
dimensioning the object in accordance with at least the one or more extracted angular relationships and the extracted line segment.