Methods and systems for hair-service based digital image searching and ranking

1 . A computerized method for implementing a hair-service based digital image search and ranking, the method comprising:

receiving, from a hair professional application, a digitalized hair service portfolio comprising a set of hair service digital images;

receiving, from the hair professional application, a set of tags for each hair service digital image of the set of hair service digital images;

automatically generating one or more annotations to each hair service digital image by:

extracting feature vectors from each hair service digital image;

aligning textual and visual information using machine translation;

learning correlations, using one or more machine learning algorithms, between training annotations and the extracted feature vectors based on the machine translating and a relevance model implementing vector space modeling; and

performing multi-class image classification on the hair service digital image to determine the one or more annotations based on the learned correlations;

indexing the set of hair service digital images according to the annotations and the tags for storing, along with the set of tags, the one or more annotations, and the extracted feature vectors in a database for vector-space retrieval;

receiving, by a digital search engine, a search of a hair service image from a customer application installed on a user device;

computing, by the digital search engine and from the received image, hair service attribute feature vectors and annotation candidates by utilizing a content-based image retrieval (CBIR) system that implements computer vision and automatic image annotation methods;

extracting, by the digital search engine, a geolocation from reverse-image analyzing the hair service image and constraining a search region based on the extracted geolocation;

searching the indexed annotations and the tags associated with the stored hair service images within the constrained search region that match the computed feature vectors and annotation candidates to generate a search result of hair service digital images;

measuring, by the digital search engine, cosine similarity between the feature vectors from the received image and the feature vectors of each stored hair service image;

obtaining, by the digital search engine, a matching hair service digital image by ranking the search result based on the measured cosine similarity; and

automatically modifying, at the user device, markup code of a mobile application interface of the customer application to display the search result on the user device, wherein the search result includes the matching hair service digital image,

wherein the customer application integrates a chat-bot service providing an interactive conversational interface via an instant messaging platform to interact with the user device through a virtual assistant.

2 . The computerized method of claim 1 , wherein the search comprises a text search.

3 . The computerized method of claim 2 further comprising:

updating the text search to include a hair service tag and a hair service color tag.

4 . The computerized method of claim 1 , wherein a machine learning model is used for computer vision training.

5 . The computerized method of claim 1 , wherein each tag comprises a hair service descriptor of a hair service depicted in the hair service digital image, and the hair service descriptor includes labeling from the hair professional application.

6 . The computerized method of claim 5 , wherein image labeling was done by a designated specialist through the hair professional application used to support training a machine learning model.

7 . The computerized method of claim 5 , wherein the hair service descriptor comprises a hair service type, a hair coloring type, or a model descriptor type.

8 . The computerized method of claim 1 , wherein the hair service portfolio comprises hair professional contact information including a location of a hair professional.

9 . The computerized method of claim 1 , wherein the matching hair service digital image is determined based on a combination of a match between the stored hair service images and the received hair service search image, and a match between a location of a hair professional with respect to a user location.

10 . The computerized system of claim 1 , wherein the search from the user further comprises a description of the user's hair type.

11 . The computerized system of claim 10 , wherein the user's hair type is obtained via a questionnaire that identifies various user hair variables, and wherein a user hair variable comprises a hair straightness value, a hair curly value, a user demographic grouping, a user age, a hair thickness value, or a hair length value.

12 . A computer system for implementing a hair-service based digital image search and ranking by a server comprising:

a processor;

a memory containing instructions when executed on the processor, cause the processor to perform operations that:

receive a digitalized hair service portfolio from a hair professional application, wherein the hair service portfolio comprises a set of hair service digital images;

receive, from the hair professional application, a set of tags for each hair service digital image of the set of hair service digital images;

automatically generate one or more annotations to each hair service digital image by:

extracting feature vectors from each hair service digital image;

aligning textual and visual information using machine translation;

learning correlations, using one or more machine learning algorithms, between training annotations and the extracted feature vectors based on the machine translating and a relevance model implementing vector space modeling; and

performing multi-class image classification on the hair service digital image to determine the one or more annotations based on the learned correlations;

index the set of hair service digital images according to the annotations and the tags for storing, along with the set of tags, the one or more annotations, and the extracted feature vectors in a database for vector-space retrieval;

receive a search of a hair service image from a customer application installed on a user device;

compute, from the received image, hair service attribute feature vectors and annotation candidates by utilizing a content-based image retrieval (CBIR) system that implements computer vision and automatic image annotation methods;

extract a geolocation from reverse-image analyzing the hair service image and constrain a search region based on the extracted geolocation;

search the indexed annotations and the tags associated with the stored hair service images within the constrained search region that match the computed feature vectors and annotation candidates to generate a search result of hair service digital images;

measure cosine similarity between the feature vectors from the received image and the feature vectors of each stored hair service image;

obtain a matching hair service digital image by ranking the search result based on the measured cosine similarity;

automatically modify, at the user device, markup code of a mobile application interface of the customer application to display the search result on the user device, wherein the search result includes the matching hair service digital image,

wherein the customer application integrates a chat-bot service providing an interactive conversational interface via an instant messaging platform to interact with the user device through a virtual assistant.

13 . The computerized system of claim 12 , wherein the search comprises a text search.

14 . The computerized system of claim 13 wherein the memory containing instructions when executed on the processor, further cause the processor to perform an operation that updates the text search to include a hair service tag and a hair service color tag.

15 . The computerized system of claim 12 , wherein a machine learning model is used for computer vision training.

16 . The computerized system of claim 12 , wherein each tag comprises a hair service descriptor of a hair service depicted in the hair service digital image, and the hair service descriptor includes labeling from the hair professional application.

17 . The computerized system of claim 16 , wherein image labeling was done by a designated specialist through the hair professional application used to support training a machine learning model.

18 . The computerized system of claim 16 , wherein the hair service descriptor comprises a hair service type, a hair coloring type, or a model descriptor type.

19 . The computerized system of claim 12 , wherein the hair service portfolio comprises hair professional contact information including a location of a hair professional, and wherein the matching hair service digital image is determined based on a combination of a match between the stored hair service images and the received hair service search image, and a match between a location of the hair professional with respect to a user location.

20 . The computerized system of claim 19 , wherein the memory containing instructions when executed on the processor, further cause the processor to perform an operation of tagging the hair service search image with a hair service tag, a hair service color tag, a hair service thickness tag, a hair service demographic tag, and a hair service length tag, wherein searching the annotations is based on tags of the received hair service search image.