Method, apparatus, and electronic devices for searching images
Methods, apparatuses and electronic devices for generating a feature vector, as well as searching methods, apparatuses and electronic devices are disclosed. The method for generating a feature vector includes: acquiring data information; extracting a semantic feature from the data information, to acquire semantic feature information; and acquiring a feature vector of the data information by using a preset function, with the semantic feature information as a parameter. The technical solution identifies picture information by recognizing semantics of image information and matching the semantics of the image information with natural language descriptions. Different from conventional image search schemes of existing search engines, this technical solution does not need to retrieve a text description of image information, but retrieves and identifies images based on the content of the image information. Therefore, results with higher accuracy may be returned compared with the existing text-based image search.
1. An apparatus comprising:
one or more processors; and
one or more memories storing thereon computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to perform acts comprising:
acquiring data information;
extracting a semantic feature from the data information, to acquire a plurality of sets of semantic feature information;
acquiring a feature vector of the data information by using a preset function, by at least summing the plurality of sets of semantic feature information;
acquiring a memorized feature variable of the plurality of sets of semantic feature information by at least averaging the sum of the plurality of sets of semantic feature information; and
updating the memorized feature variable by concatenating the memorized feature variable and the feature vector.
2. The apparatus of claim 1 , wherein the extracting the semantic feature from the data information includes:
extracting the semantic feature from the data information by using a preset neural network model.
3. The apparatus of claim 2 , wherein the preset neural network model includes ResNet-152.
4. The apparatus of claim 2 , wherein the acquiring the feature vector of the data information using the preset function, includes:
setting a function and variables of the function according to the type of the data information; and
acquiring the feature vector of the data information by using the function with the semantic feature information as the parameter.
5. The apparatus of claim 4 , wherein the data information includes text information.
6. The apparatus of claim 4 , wherein the data information includes image information.
7. The apparatus of claim 5 , wherein the extracting the semantic feature from the data information to acquire semantic feature information includes:
splitting the data information based on a feature of a preset level of the neural network model; and
acquiring each piece of split semantic feature information.
8. The apparatus of claim 7 , wherein:
the preset function includes a hyperbolic tangent function with a preset feature matrix, and a summation function; and
the acquiring the feature vector of the data information by using the preset function includes:
acquiring a sum of the semantic feature information according to the summation function;
training the sum of the semantic feature information according to the preset feature matrix, to acquire a training result; and
acquiring the feature vector of the data information according to the hyperbolic tangent function, with the training result as a parameter.
9. One or more memories storing thereon computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:
acquiring data information;
extracting a semantic feature from the data information to acquire a plurality of sets of semantic feature information;
acquiring a feature vector of the data information by using a preset function, by at least summing the plurality of sets of semantic feature information;
acquiring a memorized feature variable of the plurality of sets of semantic feature information by at least averaging the sum of the plurality of sets of semantic feature information; and
updating the memorized feature variable of the semantic feature information by concatenating the memorized feature variable and the feature vector.