1460882951-a946fe6a-65d4-499e-a108-36f80f85b667

1. A method for text filtering, the method comprising:
storing a pre-defined semantic keyword in a text filtering system, the semantic keyword comprising at least a basic keyword and a logical operator;
finding, by the text filtering system, the basic keyword of the semantic keyword in the input text according to the pre-defined semantic keyword;
in an event that a text content matching the basic keyword in the input text is found, conducting a semantic match in the found text content, the semantic match comprising matching the found text content with the semantic keyword according to the logical operator of the semantic keyword; and
in an event that the semantic match is successful, filtering a matched text context.
2. The method as recited in claim 1, wherein:
the basic keyword uses a character as a unit and is stored in the text filtering system according to a tree-type structure; and
a first character of the basic keyword is a root node and a last character of the basic keyword is a leaf node in the tree-type structure, basic keywords with an identical first character sharing a common root node.
3. The method as recited in claim 2, wherein finding the basic keyword of the semantic keyword in the input text according to the pre-defined semantic keyword comprises:
obtaining a character cl in the input text;
using c1 as a current character and using the root node of the tree-type structure as a current node to match the current character with the current node;
in an event that the current character matches the current node and the current node has a child node, matching a next character following the current character with the child node of the current node;
in an event that the current character does not match the current node and the current node has a sibling node, matching the current node with the sibling node of the current node;
connecting the current node with the root node to obtain a matching route; and
establishing the basic keyword according to the leaf node having a successful match result on the matching route.
4. The method as recited in claim 3, further comprising:
before matching the current character with the current node, determining whether the current character has a corresponding prototype character in a dictionary; and
if affirmative, converting the current character into the corresponding prototype character and using the corresponding prototype character as the current character to match with the current node.
5. The method as recited in claim 1, wherein:
the semantic keyword further comprises a filtering condition; and
the semantic match further comprises matching an attribute of the input text with the filtering condition.
6. The method as recited in claim 1, wherein:
the semantic keyword further comprises a filtering action; and
filtering the matched text context comprises filtering the matched text content according to the filtering action.
7. A text filtering system comprising:
a keyword storage unit that stores a pre-defined semantic keyword, the semantic keyword comprising at least a basic keyword and a logical operator;
a basic finding unit that finds the basic keyword of the semantic keyword in an input text according to the pre-defined semantic keyword;
a semantic matching unit that, in an event that a text content matching the basic keyword in the input text is found, conducts a semantic match in the found text content, the semantic match comprising matching the found text content with the semantic keyword according to the logical operator of the semantic keyword; and
a filter processing unit that, in an event that the semantic match is successful, filters a matched text context.
8. The system as recited in claim 7, wherein:
the keyword storage unit uses a character as a unit and stores the basic keyword according to a tree-type structure; and
a first character of the basic keyword is a root node and a last character of the basic keyword is a leaf node, basic keywords having an identical first character sharing a common root node.
9. The system as recited in claim 7, wherein the basic finding unit comprises:
a text obtaining sub-unit that obtains a character c1 in the input text;
a character matching sub-unit that uses c1 as a current character and uses the root node of the tree-type structure as a current node to match the current character with the current node,
in an event that the current character matches the current node and the current node has a child node, the character matching sub-unit matching a next character following the current with the child node of the current node,
in an event that the current character does not match the current node and the current node has a sibling node, the character matching sub-unit matching the current node with the sibling node of the current node; and

a determination sub-unit that connects the current node and the root node to obtain a matching route, and establishes the basic keyword according to the leaf node having a successful match result on the matching route.
10. The system as recited in claim 9, wherein:
the basic finding unit further comprises a character conversion sub-unit that determines whether the current character has a corresponding prototype character in a dictionary before the character matching sub-unit conducts matching and, if affirmative, converts the current character into the corresponding prototype character; and
the character matching sub-unit uses the corresponding prototype character as the current character to match with the current node.
11. The system as recited in claim 7, wherein:
the semantic keyword comprises a filtering condition; and
the semantic matching unit comprises a category matching sub-unit that matches an attribute of the input text with the filtering condition.
12. The system as recited in claim 7, wherein:
the semantic keyword comprises a filtering action; and
the filter processing unit filters the matched text content according to the filtering action.
13. One or more computer-readable storage media having computer-executable instructions stored thereupon that, when executed by a computer, cause the computer to carry out a process, the process comprising:
storing a pre-defined semantic keyword in a text filtering system, the semantic keyword comprising at least a basic keyword and a logical operator;
finding, by the text filtering system, the basic keyword of the semantic keyword in the input text according to the pre-defined semantic keyword;
in an event that a text content matching the basic keyword in the input text is found, conducting a semantic match in the found text content, the semantic match comprising matching the found text content with the semantic keyword according to the logical operator of the semantic keyword; and
in an event that the semantic match is successful, filtering a matched text context.
14. The one or more computer-readable storage media as recited in claim 13, wherein:
the basic keyword uses a character as a unit and is stored in the text filtering system according to a tree-type structure; and
a first character of the basic keyword is a root node and a last character of the basic keyword is a leaf node in the tree-type structure, basic keywords with an identical first character sharing a common root node.
15. The one or more computer-readable storage media as recited in claim 14, wherein finding the basic keyword of the semantic keyword in the input text according to the pre-defined semantic keyword comprises:
obtaining a character c1 in the input text;
using c1 as a current character and using the root node of the tree-type structure as a current node to match the current character with the current node;
in an event that the current character matches the current node and the current node has a child node, matching a next character following the current character with the child node of the current node;
in an event that the current character does not match the current node and the current node has a sibling node, matching the current node with the sibling node of the current node;
connecting the current node with the root node to obtain a matching route; and
establishing the basic keyword according to the leaf node having a successful match result on the matching route.
16. The one or more computer-readable storage media as recited in claim 15, further comprising:
before matching the current character with the current node, determining whether the current character has a corresponding prototype character in a dictionary; and
if affirmative, converting the current character into the corresponding prototype character and using the corresponding prototype character as the current character to match with the current node.
17. The one or more computer-readable storage media as recited in claim 13, wherein:
the semantic keyword further comprises a filtering condition; and
the semantic match further comprises matching an attribute of the input text with the filtering condition.
18. The one or more computer-readable storage media as recited in claim 13, wherein:
the semantic keyword further comprises a filtering action; and
filtering the matched text context comprises filtering the matched text content according to the filtering action.

The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.

1. A method for multimedia model retrieval, comprising:
acquiring parameters of a first multimedia model to be retrieved;
performing a projection on the first multimedia model according to the parameters of the first multimedia model so as to obtain a projection image;
performing a feature extraction on the projection image to generate a first feature extraction result;
comparing the first feature extraction result with a plurality of feature extraction results corresponding to a plurality of multimedia models stored in a database so as to obtain a retrieval result, wherein the retrieval result comprises a number of multimedia models;
labeling by a user each of the multimedia models in the retrieval result with a probability value;
training a support vector machine with the labeled multimedia models;
performing a probability-based classification on the plurality of multimedia models stored in the database by using the support vector machine to obtain a classification result; and
updating the retrieval result with the classification result.
2. The method according to claim 1, wherein the performing the feature extraction on the projection image comprises: performing the feature extraction on the first multimedia model by using a Zernike moment.
3. The method according to claim 1, further comprising: performing a pre-processing on the parameters of the first multimedia model to be retrieved.
4. The method according to claim 2, further comprising: performing a pre-processing on the parameters of the first multimedia model to be retrieved.
5. The method according to claim 3, wherein the pre-processing comprises:
performing an isotropic transformation on the first multimedia model; and
performing a vertex sampling according to the first multimedia model after the isotropic transformation.
6. The method according to claim 1, wherein the performing a probability-based classification on the plurality of multimedia models stored in a database by using the support vector machine to obtain a classification result comprises:
forecasting each of the multimedia models in the database, obtaining relevant or irrelevant probability values attributed to each of the multimedia models in the database and revising a similarity degree by the probability values.
7. The method according to claim 1, wherein the performing the projection on the first multimedia model according to the parameters of the first multimedia model comprises: performing a spherical projection on a three-dimensional multimedia model and calculating a spherical projection image.
8. The method according to claim 1, further comprising: sorting the retrieval result obtained by comparing or the retrieval result obtained after updating.
9. The method according to claim 1, wherein a file format of the parameters of the first multimedia model to be retrieved is a three-dimensional model file format .3ds or a text file format .txt.
10. The method according to claim 1, wherein the probability-based classification comprises a similarity classification which is based on posterior probability.
11. An apparatus for multimedia model retrieval, comprising: at least one hardware processor operable to:
acquire parameters of a first multimedia model to be retrieved;
generate a projection image from the first multimedia model according to the parameters of the first multimedia model;
perform a feature extraction on the projection image to generate a first feature extraction result;
compare the first feature extraction result with a plurality of feature extraction results corresponding to a plurality of multimedia models stored in a database so as to obtain a retrieval result, wherein the retrieval result comprises a number of multimedia models; and
train a support vector machine with multimedia models labeled by a user upon the retrieval result with a probability value, and perform a probability-based classification on the plurality of multimedia models stored in a database by the support vector machine to obtain a classification result, and update the retrieval result with the classification result.
12. The apparatus according to claim 11, wherein the at least one hardware processor is further operable to perform a pre-processing on the parameters of the first multimedia model to be retrieved.
13. The apparatus according to claim 12,
wherein the at least one hardware processor is further operable to perform an isotropic transformation on the first multimedia model; and
to perform a vertex sampling on the first multimedia model after the isotropic transformation.
14. The apparatus according to claim 11,
wherein the at least one hardware processor is further operable to employ the multimedia models manually labeled by a user upon the retrieval result with a probability value as a training sample and train according to the training sample to obtain the support vector machine; and
to perform a probability-based classification on the plurality of multimedia models stored in a database by the support vector machine to obtain a classification result, and update the retrieval result with the classification result.
15. The apparatus according to claim 11, wherein the at least one hardware processor is further operable to sort the retrieval result obtained by comparing or the retrieval result obtained after updating.
16. A system for multimedia model retrieval, comprising:
a database management system adapted to store multimedia model file information;
a model retrieval apparatus adapted to acquire parameters of a first multimedia model to be retrieved, perform a projection on the first multimedia model according to the parameters of the first multimedia model so as to obtain a projection image, perform a feature extraction on the projection image to generate a first feature extraction result, comparing the first feature extraction result with a plurality of feature extraction results corresponding to a plurality of multimedia models stored in the database management system to obtain a retrieval result, wherein the retrieval result comprises a number of multimedia models, train a support vector machine with the multimedia models labeled by a user upon the retrieval result with a probability value, perform a probability-based classification on the plurality of multimedia models stored in a database by using the support vector machine to obtain a classification result and update the retrieval result with the classification result.