The image-based search described herein is not limited to online dating " Imlooking: Image-based Face Retrieval in Online Dating Profile Search," CHI Instead of directly measure the difference between two images, we build a feature X.: Imlooking: image-based face retrieval in online dating profile search . I used to think of going on a first date like prepping for a job But what do you do when you're faced with a message on a dating site and all you've got is their screenname, Taking their photo and plugging it into a reverse image search . be explained pretty simply: your brain's going into recovery mode.
Memory may include an operating systemone or more application programs for implementing all or a part of the image-based search, as well as various other data, programs, media, and the like.
In one implementation, the memory includes an image-search application including a user interface modulea data management moduleand a search module The search module interacts with the user interface module and data storage module to perform search functions, such performing textual searches using conventional text search methodologies, comparing query images to stored images in, for example, the stored image database Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.
Memoryremovable storage and non-removable storage are all examples of computer storage media. Additional types of computer storage media that may be present include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks DVD or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by the image search server or other computing device.
Communications connection s is an example of communication media. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared and other wireless media. The term computer readable media as used herein includes both storage media and communication media.
The image search server may also include input device s such as a keyboard, mouse, pen, voice input device, touch input device, etc.
All these devices are well know in the art and need not be discussed at length here. Exemplary User Interface illustrate an exemplary user interface usable with the image-based search of FIG.
Specifically, the query interface includes a text-based portionwhich allows the user to enter text-based non-visual search criteria, such as gender, age, and location of potential companions to search for. The text-based portion is shown as a structured search form, but could additionally or alternatively include one or more freeform search fields. The text-based portion is provided for its ease and effectiveness at accurately recognizing age, gender, ethnicity, location and other data which is readily recorded in text data.
However, in some implementations, the text-based portion of the query interface may be optional or may be omitted entirely. The query interface also includes an image-based search portionwhich allows the user to provide a query image to use to search for similar images. In the described implementation, the query image of the face should be a near-frontal view.
The query image may be automatically aligned and cropped to include only the face area. The cropped image may then be displayed in the left part of the image-based search portion of the query interface Since at least some people have clear preferences for certain facial features, the image-based search portion also may include a feature preference inputwhich allows a user to input a preference for one or more facial features of the query image. In the implementation of FIG. The user may indicate his or her preference by filling the appropriate number of boxes on the scale associated with each of those facial features.
For example, in FIG. Categorized search results are displayed to the user, including one or more resultant images and associated profile information e. In the example shown in FIG.
Each category corresponds to a row of the search resultswith three resultant images being displayed for each category.
The categories are laid out with the holistic similarity first, followed by the facial features in the order of the preference weight specified by the user on the query interface Thus, because the user in FIG. However, the results could be displayed in any other configuration or order, and could be categorized into any number of different groups based on facial features, age, ethnicity, location or any other available criteria.
Also, the user may run a new search using any of the resultant images as the new query image for the search by simply selecting a similar face search button associated with the respective resultant image to be used as the new query image. The method may, but need not, be implemented at least partially by an image-based search system such as that shown in FIG.
The method comprises, atprompting a user to enter profile information about the user to set up the user's account. The profile information may include textual information including the user's gender, age, ethnicity, and location, as well as one or more pictures of the user. The profile information of the user is then stored, atin one or more databases.
In one example, the textual information is stored in one database, while the picture s of the user are stored in another database. Alternatively, the textual information and pictures may be stored together in the same database. Atthe user is prompted to enter a textual description of one or more characteristics of a person for whom they would like to search.
The textual description may be a structured or freeform text search, and may include any characteristic that is readily recordable in textual format. Some examples of information that can readily be recorded in textual format include gender, age, ethnicity, location, height, weight, and the like. Atthe user is prompted to provide a query image of a face for which they would like to search for similar faces. The query image may be provided by the user uploading or otherwise inputting an image from the user's own collection, by selecting an image from among the stored images to use as the query image, or by selecting an image from the Internet to use as the query image.
The user is then prompted, atto indicate a preference for one or more facial features of the query image. The user's preference may be indicated by assigning a weighted preference value to one or more facial features, such as the eyes, nose, mouth, and face shape of the query image.
However, numerous other ways of indicating a preference value may additionally or alternatively be used.
Zoosk review: Easy to use and a great design, but can get a bit spammy
Also, any number and type of facial features may be assigned preference values. Additional facial features that could be assigned values include hair, eyebrows, ears, chin, cheeks, or any other perceptible facial features. Atthe database of stored images is searched for images of faces corresponding to users having the characteristics specified by the textual search at The subset of stored images matching the textual search may be returned to the user, or the subset of stored images may be used as the field of search for the image-based portion of the search.
That is, the textual search could be performed prior to the image-based search to narrow the field of images to search in the image-based search.
In that case, atfeatures of the query image face and faces in the subset of stored images are detected and extracted for feature-based analysis.
- US7684651B2 - Image-based face search - Google Patents
As discussed above, the survey may be conducted ahead of time or may be generated and updated in real-time. The mapping function can be generated in any known manner, such as calculating pair-wise difference vectors in the feature space of a plurality of stored images, and learning the mapping function for each pair of images using a SVM machine, as described above, or using any of a variety of other known techniques. The difference vectors can then be mapped to the average score of the human assessors.
I asked Tinder for my data. It sent me 800 pages of my deepest, darkest secrets
The search results are displayed to the user at Specifics of one exemplary search method are described above. For example, the text-based search may be conducted independently of the image-based search, or may be omitted entirely, if so desired. Moreover, any of the acts described above may be implemented by a processor or other computing device based on instructions stored on one or more computer-readable media associated with the image-based search system or a component thereof.
As discussed above, computer-readable media can be any available media that can be accessed by the image-based search system or a component thereof. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claims.
For example, the methodological acts need not be performed in the order or combinations described herein, and may be performed in any combination of one or more acts.
The one or more computer-readable media of claim 1wherein prompting the user to indicate a preference for one or more facial features comprises prompting the user to assign a weighted preference value to the one or more facial features of the query image of the face. The one or more computer-readable media of claim 1wherein prompting the user to indicate a preference for one or more facial features comprises prompting the user to indicate one or more of the facial features of the query image of the face as being more preferred than one or more other facial features.
The one or more computer-readable media of claim 1wherein the one or more facial features comprise eyes, mouth, nose, and shape of the face in the query image. The one or more computer-readable media of claim 1wherein prompting the user to indicate a preference for one or more facial features comprises prompting the user to assign a weighted preference value for each of the eyes, mouth, nose, and shape of the face in the query image.
The one or more computer-readable media of claim 1wherein displaying the resultant images comprises clustering the resultant images into multiple groups based on facial features. The algorithm will learn everything about you and your likes and dislikes as you use the app.
USB2 - Image-based face search - Google Patents
Luckily, there is a profile verification system that helps real users, who are indicated with a green check mark, to weed out the not-so-genuine ones. It will prompt you with three ways to verify your account, by photo, phone number, or Twitter account. If you select photo verification, the app will give you a few instructions. Your photo is then sent to administrators for verification. This process took a few hours to complete, instead of a few seconds with a phone number it sends you a special PIN via text message and Twitter verification.
Zoosk does the work to ensure that your "video" photo lines up with the one on your profile. Although it takes some time, you can now rest easy knowing your matches are real or at least verified when you see that green check mark on their photo. And in that way, the online dating service is really easy to spam people with shady and phony users who advertise for free sex if you follow a link to another website or dating app.
The green check verification does come in handy in those situations, but it can be tough to scroll through all the accounts to seek them out. Within one minute of creating a new profile, I was messaged eight times from people who seemed not to be real. Most of them simply viewed my new profile, but some wanted to meet or sent a generic message to elicit a response. You may run into another problem with seemingly inactive users. Out of the 11 messages and notifications I received during my review period, it was difficult to tell which ones were genuine and which ones were not.
There are still a number of real people on Zoosk who are excited to match with you. The design of the website and app are pretty modern, as they both emphasize engagement and interaction. In fact, the first thing you do after you create an account is start "liking" people, so the algorithm can get a feel of your preferences. You can either like someone with a smiley face, which indicates friendship, or you can like someone with a heart, which indicates love.
Refine your matches Zoosk lets you cast as wide or as small of a net as you please, with preferences that you can broaden or refine. You have the chance to match with people depending on their location between three miles towhile the default setting is "auto-selected.PLENTY OF FISH MESSAGES: 3 Openers & Text Examples To Get More Girls
You can also go a bit deeper with settings for height, religion, relationship history, body type, children, ethnicity, education, and smoking preferences all in the mix to tweak or leave alone. It seems no matter what your type, the platform has someone in mind for you. Is Zoosk worth its membership price?
With a basic account, you can only browse and like other users. While other sites like OkCupidTinder, and Bumble have a robust free option that allows you to read, send, and receive messages, Zoosk does not.