Mixing segmentation algorithms utilizing soft classifications to identify segments of three-dimensional digital models

Patent metadata

Field	Value
Publication number	US-11315255-B2
Application number	US-202016907663-A
Country	US
Kind code	B2
Filing date	Jun 22, 2020
Priority date	Apr 14, 2017
Publication date	Apr 26, 2022
Grant date	Apr 26, 2022

Abstract

The present disclosure includes methods and systems for identifying and manipulating a segment of a three-dimensional digital model based on soft classification of the three-dimensional digital model. In particular, one or more embodiments of the disclosed systems and methods identify a soft classification of a digital model and utilize the soft classification to tune segmentation algorithms. For example, the disclosed systems and methods can utilize a soft classification to select a segmentation algorithm from a plurality of segmentation algorithms, to combine segmentation parameters from a plurality of segmentation algorithms, and/or to identify input parameters for a segmentation algorithm. The disclosed systems and methods can utilize the tuned segmentation algorithms to accurately and efficiently identify a segment of a three-dimensional digital model.

First claim

We claim:

1. A computer-implemented method for manipulating segments of three-dimensional digital models, comprising:
determining a soft classification of a three-dimensional digital model utilizing a classification algorithm;
generating segmentation parameters for the three-dimensional digital model utilizing multiple segmentation algorithms;
generating combined segmentation parameters by combining the segmentation parameters from the multiple segmentation algorithms utilizing the soft classification of the three-dimensional digital model; and
identifying a segment of the three-dimensional digital model utilizing the combined segmentation parameters.

2. The computer-implemented method of claim 1 , further comprising determining the soft classification of the three-dimensional digital model by:
determining a first probability of a first soft classification and a second probability of a second soft classification utilizing the classification algorithm; and
combining the segmentation parameters utilizing the first probability and the second probability.

3. The computer-implemented method of claim 2 , wherein combining the segmentation parameters utilizing the first probability and the second probability comprises:
utilizing the first probability to determine a first mixture coefficient for a first set of segmentation parameters generated utilizing a first segmentation algorithm;
utilizing the second probability to determine a second mixture coefficient for a second set of segmentation parameters generated utilizing a second segmentation algorithm; and
combining the first set of segmentation parameters and the second set of segmentation parameters utilizing the first mixture coefficient and the second mixture coefficient.

4. The computer-implemented method of claim 2 , further comprising utilizing the first probability and the second probability to combine the segmentation parameters in response to determining that the first probability and the second probability satisfy a probability threshold.

5. The computer-implemented method of claim 1 , further comprising:
selecting a first input parameter of a first segmentation algorithm of the multiple segmentation algorithms based on the soft classification; and
selecting a second input parameter for a second segmentation algorithm of the multiple segmentation algorithms based on the soft classification.

6. The computer-implemented method of claim 5 , further comprising:
utilizing the first input parameter of the first segmentation algorithm to generate a first set of segmentation parameters of the segmentation parameters; and
utilizing the second input parameter of the second segmentation algorithm to generate a second set of segmentation parameters of the segmentation parameters.

7. The computer-implemented method of claim 1 , wherein identifying the segment of the three-dimensional digital model comprises:
receiving a user selection of a portion of the three-dimensional digital model; and
determining a selected segment corresponding to the user selection utilizing the combined segmentation parameters.

8. The computer-implemented method of claim 1 , wherein generating the segmentation parameters comprises generating a first set of edge segmentation scores utilizing a first segmentation algorithm and generating a second set of edge segmentation scores utilizing a second segmentation algorithm.

9. A non-transitory computer readable medium storing instructions, that when executed by at least one processor, cause a computer system to:
determine a soft classification of a three-dimensional digital model utilizing a classification algorithm;
generate a first set of edge segmentation scores for the three-dimensional digital model utilizing a first segmentation algorithm;
generate a second set of edge segmentation scores for the three-dimensional digital model utilizing a second segmentation algorithm;
generate a combined set of edge segmentation scores by combining the first set of edge segmentation scores and the second set of edge segmentation scores based on the soft classification; and
segment the three-dimensional digital model utilizing the combined set of edge segmentation scores.

10. The non-transitory computer readable medium of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the computer system to:
select a first input parameter of the first segmentation algorithm based on the soft classification; and
generate the first set of edge segmentation scores utilizing the first input parameter and the first segmentation algorithm.

11. The non-transitory computer readable medium of claim 10 , further comprising instructions that, when executed by the at least one processor, cause the computer system to:
select a second input parameter of the second segmentation algorithm based on the soft classification; and
generate the second set of edge segmentation scores utilizing the second input parameter and the second segmentation algorithm.

12. The non-transitory computer readable medium of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the computer system to determine the soft classification of the three-dimensional digital model by:
determining a first probability of a first soft classification utilizing the classification algorithm; and
determining a second probability of a second soft classification utilizing the classification algorithm.

13. The non-transitory computer readable medium of claim 12 , further comprising instructions that, when executed by the at least one processor, further cause the computer system to:
utilize the first probability to determine a first mixture coefficient for the first set of edge segmentation scores;
utilizing the second probability to determine a second mixture coefficient for the second set of edge segmentation scores; and
mix the first set of edge segmentation scores and the second set of edge segmentation scores by applying the first mixture coefficient to the first set of edge segmentation scores and applying the second mixture coefficient to the second set of edge segmentation scores.

14. The non-transitory computer readable medium of claim 9 , further comprising instructions that, when executed by the at least one processor, further cause the computer system to:
identify a correspondence between the soft classification, the first segmentation algorithm, and the second segmentation algorithm; and
in response to determining the soft classification of the three-dimensional digital model utilizing the classification algorithm, selecting the first segmentation algorithm and the second segmentation algorithm from a plurality of segmentation algorithms based on the correspondence.

15. A system comprising:
one or more memory devices; and
one or more servers that are configured to cause the system to:
determine a soft classification for a three-dimensional digital model utilizing a classification algorithm;
generate segmentation parameters for the three-dimensional digital model utilizing multiple segmentation algorithms;
generate combined segmentation parameters by combining the segmentation parameters from the multiple segmentation algorithms utilizing the soft classification of the three-dimensional digital model; and
identify a segment of the three-dimensional digital model utilizing the combined segmentation parameters.

16. The system of claim 15 , wherein the one or more servers are further configured to select the multiple segmentation algorithms from a plurality of se

Assignees

• Adobe Inc

Inventors

• Kim Vladimir
• Hertzmann Aaron
• Yumer Mehmet

Classifications

• G06V10/26

  Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion · CPC title

• G06V30/248

  involving plural approaches, e.g. verification by template match; Resolving confusion among similar patterns, e.g. "O" versus "Q" (G06V30/242 takes precedence) · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06V20/64

  Three-dimensional [3D] objects · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title