How to find the place where the photo was taken. Finding out a person’s location via VK Where are the photos located?

The search giant's new development has already surpassed humans.

Most of you know about Google's photo search feature. Just go to Pictures mode, add a link to a photo or upload a file from your computer, and the search engine will identify the object located in the frame.

However, this trick only works with popular attractions and famous places. Google can easily recognize the Eiffel Tower, the Statue of Liberty or the Egyptian pyramids. However, if you simply upload a photo of a street in your city, the search will not be successful.

Experts at Google are not sitting still and developing a neural network PlaNet, which will allow you to search for an area using photos more accurately. For analysis and search, this system divided most of the land into 26,000 zones. Each of them is characterized by certain features and differs from the others.


More than 120 million images with added geo-tags were used to train PlaNet. Later, to test the system, she was asked to recognize the location of 2 million photos from Flickr. The neural network was able to determine the position 3,6% street-accurate images, 10% - accurate to the city. The country was set to 28,4% cases, and the continent - in 48% .


PlaNet notices something in the photo that not every person will notice. The system recognizes dishes, plants, direction of movement on the road, animals, architecture of buildings and other features. After analysis, the search engine makes its guess about the shooting location.

PlaNet's results were compared with the results of travel experts in GeoGuessr. This game uses Google Street View street photos and requires you to guess the location where the photo was taken.

The neural network performed better than a team of experts. The average PlaNet error result was 1131.7 km, travelers were wrong by an average of 2320.75 km.

I wonder if PlaNet will be able to guess the location of a photo based on the carpet in the background.

In this material I will tell you how to search by photo on social networks, and what we need for this. Imagine a situation: you have a photo of a person about whom you need to find all the available information. Social networks, which often contain accounts of the people we need, can provide invaluable assistance in the process of such a search. They may contain a variety of information about a person, including his photos and videos, interests, hobbies, favorite books and films, and other data useful to us. But how can you find such a page with just a photo?

What methods of searching for a person are available on the Internet?

You can search for the right person on social networks in various ways (by first and last name, place of residence, geotags, age, and other relevant descriptors). If you already have a photograph of him, then you can turn to the capabilities of special network tools that can provide significant assistance in our search.

Some of these resources (for example, the popular “FindFace”) are specially “tailored” to search for people in a specific network (in the case of “FindFace” this is the Russian “VK”). Others, at the level of Google and Yandex search engines, allow you to search for identical and similar images on many sites (including social networks), displaying on the screen all the richness of the results obtained.

In some cases, it is necessary to prepare the desired photo to search for it (for example, using the template graphic editor “”). It is necessary to get rid of the presence of other people in the photo (if any), as well as remove other unnecessary details that could interfere with the search service in its work.

Let's look at ways to search for a person by their face image on social networks.

“FindFace” - effective photo search using neural networks

In November 2015, the University of Washington organized the “The MegaFace Benchmark” competition, in which more than a hundred teams took part, including from Google and Facebook. The goal of the competition is to create the most effective face recognition algorithm, allowing you to accurately select a similar face among many analogues.

Despite strong competition, the competition was won by the Russian company N-Tech.Lab. Its algorithm, built on the work of neural networks, provided the highest recognition accuracy among analogues - about 75% of all results obtained.

• Later, this algorithm was embodied in the findface.ru service, which allows you to find the right person (or people similar to him) on the Vkontakte social network.
• At the same time, a mobile application of the same name is also available to Android OS users.

The “findface” service will allow you to quickly find the right person on VK

How to use Fifindface search

How does this service work? To work with it, first of all you need to have a page on VK (if you don’t have one, create one). Then do the following:

1. Go to findface.ru;
2. Click on the red “Find identical” button;
3. Allow the service to access your account by clicking on “Allow”;
4. Upload the photo you need to search and wait for the search procedure to complete;
5. Review your results.

The free capabilities of the service are limited to a certain number of searches on the social network (usually several dozen). For unlimited possibilities, you will have to pay the developers.

https://youtu.be/do9cdiD9e8E

“Google Images” – with the ability to find a person by photo

The capabilities of the popular international search engine “Google” allow you to search by image online, including searching by photo of a person on social networks. In this case, the search engine displays both identical and similar results, with information on the found image (including the probable first and last name of the person in the photo).

To work with the service from Google, do the following:

"Yandex.Pictures" - will find anyone on a social network based on a face picture

Image search on social networks from the Yandex service is organized in a manner similar to Google. To search, go to “Yandex.Pictures” or use to launch the service, click on the camera image on the right, upload the desired picture to the resource (or provide a link to it on the Internet). Review your results.

You can also use a special Yandex tool that searches for people on social networks. Follow this link https://yandex.ru/people, enter the first and last name of the desired person in the search bar, click on “Find”, and view the result (usually a photo, a link to a social profile, short information).

To search, go to “Yandex.Images” and click on the camera icon on the right

Alternative resources for searching people by image

Among other services that search for images on social media. networks, I would also note the following resources:

Conclusion

Search by photo on social media networks will help you perform the services I listed above. Among the services mentioned, I recommend paying special attention to the findface.ru service - its recognition efficiency is at a fairly high level, allowing you to quickly find the pages of the people you need on VK.

Thanks to the availability of special web services and applications, you can easily find a person by photo.

Most modern programs and websites use artificial intelligence technology.

It allows you to increase the accuracy of facial recognition in a photo as the number of program users increases.

It should be understood that no search engine can give a 100% guarantee of finding an image. Often errors occur when reading pixels, or the browser searches for photos that are similar in meaning and color palette.

Below are some of the most effective ways. If one of them does not help you, feel free to move on to the other.

Ultimately, one of the methods will turn out to be the most suitable for your picture and the person will be found. We also recommend that you carry it out using several photographs at once, if you have them.

ServiceGoogle

The first and easiest way to search by image is the well-known Google. The developers of the popular system have introduced a function for working with user images.

Now you can not only enter a query, but also insert individual images into the text field. The search engine carries out analysis for all sites and social networks that are indexed in the system.

Let's look at a simple example of how to use the function:

• Let's go to the website. This page differs from the standard one by the presence of an additional button for adding;

• To get started, click on the camera icon. A window for selecting the type of photo upload will appear. The user can provide a direct link to an image already posted on the Internet or download graphics from the computer’s memory;
• Click on the “Search by image” button.

The result of processing your request in Google will look like this:

It will automatically indicate the person's name or subject. Also, the user will be able to view similar images. Another element of search results are pages that mention the person found.

Often, the result only shows similar images without providing a name. This means that there are too many similar pictures on the Internet and the system cannot find a 100% exact match. Try scrolling down to the "Pages with matching images" box. This way you can view a list of links that contain what you are looking for.

With it you can search not only for people, but also for various objects. As practice shows, a search engine can be very useful when you need to find out the name of an unknown item or find things you like.

Yandex

Photo search from Yandex is another popular service that searches for graphic content across all sites. At first glance, it is practically no different from Google, except for the interface and design. However, Yandex developers claim that their system uses a modified algorithm using computer vision.

This way, by using both systems, you can have a better chance of finding the one you're looking for. To start the analysis, go to the main site. Find the text line and next to it click on the “Pictures” tab to go to the corresponding section:

• In the tab that opens, click on the camera icon;

• Now in the tab that opens, click on the button to load a photo from the PC memory or provide a direct link to the picture on the Internet. Click on “Find”;

The result will be a set of the most similar images with which you can find a person. Also, you can select different picture sizes and continue the analysis with a better quality photo. Be sure to look at the list of mentions on other sites to increase your chances of discovering the person.

Analysis by photo inVK

In addition to standard systems, there are a lot of services for monitoring by photo only in specific social networks. Since 95% of Runet users have a profile on VKontakte, it is better to search for it using special resources that work only with this social network. This way, you will reduce the analysis time and the number of similar pictures.

To start monitoring photos on VK, you need to install an extension for the Chrome browser. The app is called "StopFake". Go to the software store and enter the name of the extension. Then click on the install key and wait for the button to appear in the upper right corner.

Although the program is designed to quickly identify fake accounts, it can also be used to find people. You will need to upload a photo to your album on VKontakte, since this can only be done using the one that is in the site’s database.

• Open the photo in your album in VK. The album can be private, the main thing is that it is on the website;
• After opening the image, press the key to launch the StopFake extension. It is located in the upper right corner of the browser toolbar. To eliminate possible errors in the operation of the utility, we recommend restarting the browser immediately after installing the extension;
• It will automatically identify them on the social network that contains the photo you are looking for. Thanks to the author's signature, you can easily find the person you need. Click on the author's name to go to it.

Tweet

Let's look for a higher resolution option. Open Google Images and click the inconspicuous button in the line:

We will have a choice - provide a link to the photo or upload it from your computer. I have a picture on my hard drive, so I click Upload file - Select file - Open. After some time (the image should be uploaded to Google’s servers), a page with search results will appear:

To filter pictures by quality, there are two links - “Small” and “Large”. Naturally, I will choose " Large" to see the best quality:

The picture resolution is very good - 1920 by 1080 pixels. So you can follow the link and download it for yourself.

So, to find a picture in better quality than it is, you need to specify its address or download it from disk, then click the “Large” link.

Searching for a person on the Internet by photo

A rather unobvious way to use the image search service. Let's say we have a photograph of a person. If the same photo or a very similar one is posted somewhere on social networks, there is a chance that a Google search engine will find it. For example, this is what will happen if you upload one of my photos:

Google decided that my photo was similar to photos of Japanese athletes. This is funny and not true, but the search was successful - we saw a link to my VKontakte page, where there is this photo, and also found out my first and last name.

Of course, such luck does not always occur. Google Images can't find photos from private albums. Otherwise, this is a great way to determine the authenticity of a photo and/or determine who is in it.

Find out the name of a thing or landscape from a photograph

There is a program called Google Goggles for iPhone and Android phones. It allows you to translate photographed text or determine the name, find information about a photographed object or landmark. Really great stuff, but there is no version for computers. This can be partially compensated for using the same Google image search.

Example: we have a photo of a landscape, but we don’t know what kind of beautiful place it is:

As in the previous options, click the button in the form of a camera in the search bar, upload a photo and get the result:

I warn you - this doesn't always work. For the correct result, this photo or a very similar one must already be in the Google image search database, i.e. be on the Internet, there should also be a description next to the picture on the page. If sights can be easily recognized (since a famous place is “known” on the Internet), then with objects the chance of finding a description is almost zero.

If there is no information about the place, we will see a page with similar photos, similar to the following example.

Previously, the TinEye search engine was strong in this, as I said earlier. But Google Images is also not perfect.

And click the link “ Similar images". Let me remind you: you can not upload a photo, but indicate its address on the Internet, or simply enter the query “red kitten” into the search, hover your mouse over the photo and click “ Similar«.

In any case, these actions will lead to a search for similar photos:

Pay attention to the inscription “Similar” above the pictures. This means that this mode is working now. Clicking the cross on the right will turn it off. Clicking on the picture will open, as it should, an enlarged copy in the background of the site where this picture is located.

Conclusion

Google Image Search always works the same way. Depending on what was found, the search page changes. If the result is very successful, when it is possible to accurately identify the name of the photo and find information on it, we are redirected to a text search page (as in the case of searching for a landmark).

In fact, the search site is trying to predict what we really need. Sometimes successful, sometimes not. There is virtually no difference between the examples above. The search always happens the same way, we just look in different search places and click the links we need. I only gave the most “contrasting” examples, so you should not look for the “Search for similar photos” or “Search for a person by photo” button. There is none of them. The search is always the same. It all depends on what we are looking for.

Google Image Search is a very powerful tool with many settings and features. Finding a similar picture or page on a social network from a person’s photo is possible. Of course, there are many “buts” here, because the search is carried out by complex but predictable algorithms, and not by real intelligence. In any case, the ability to use image search is a useful skill that everyone who uses the Internet should master.

Let's say you have some kind of image (drawing, picture, photograph), and you want to find the same (duplicate) or similar one on the Internet. This can be done using special tools from Google and Yandex search engines, the TinEye service, as well as the amazing PhotoTracker Lite browser extension, which combines all these methods. Let's look at each of them.

Search by photo in Google

1. Provide a link to the image on the Internet
2. Uploading a file from a computer

As a result, we get a complete list of similar pictures from the image that was chosen as a sample:

There is another good way that works in the Chrome browser. While on the page with the image you are interested in, move the mouse cursor to it, right-click and in the tooltip that opens, select “Find image (Google)”:

You will be immediately taken to the search results page!

Search by images in Yandex

Enter the address of the image on the Internet or upload it from your computer (you can simply drag it to a special area at the top of the browser window):

The search result looks like this:

You instantly have access to the following information:

• What are the online dimensions of the image you uploaded as a sample for searching?
• List of sites where it appears
• Similar pictures (modified based on the original one or based on which the algorithm decided on their semantic similarity)

Many have probably already heard about the online service TinEye, which Russian-speaking users often call Tinai. It is developed by experts in the field of machine learning and object recognition. As a consequence of all this, Tinay is great not only for finding similar pictures and photographs, but their components.

TinEye's indexed image database contains more than 10 billion items, and is the largest in the entire Internet. “Everything can be found here” - this phrase perfectly characterizes the service.

There is another way to search in one click. By default, the “Show quick search icon” item is activated in the application settings. When you point at a photo or picture, a round green icon pops up, clicking on which starts a search for similar images - search results for Google, Yandex, Tinay and Bing will automatically open in new tabs.

The extension was created by our compatriot, whose hobbies are closely related to photography. He originally created this tool to quickly find his photos on other people's sites.

When you might need it

• You are a photographer, you post your photos on the Internet and want to see on which sites they are used and where your copyrights may be violated.
• You are a blogger or copywriter, write articles and want to choose an “unhackneyed” image for your material.
• What if someone uses your photo from your VKontakte or Facebook profile as an avatar on a forum or a fake account on some social network? But this is more than possible!
• You found a photo of an actor you know and want to remember his name.

In fact, there are a huge number of cases when searching by photo can be useful. You can give another example...

How to find the original of a given image

For example, you have some kind of photograph, perhaps cropped or photoshopped, and you want to find its original, or a better quality version. How to do it? Conduct a search in Yandex and Google, as described above, or using PhotoTracker Lite and get a list of all images found. Next, follow the following:

1. The original image is usually larger and of better quality than the altered copy resulting from cropping. Of course, you can set a picture to any size in Photoshop, but when you enlarge it relative to the original, artifacts will always be observed. They can be easily noticed even with a cursory visual inspection.
2. Original photographs often have watermarks indicating the author of the photo (last name, website address, company name, etc.). Of course, anyone can add a watermark to absolutely any image, but in this case, you can look for a sample photo on the website or by the author’s last name, he probably posts his portfolio online somewhere.
3. And finally, a very simple sign. If your sample photo is black and white (sepia, etc.), and you find the same, but full-color photo, then you obviously do not have the original. much more difficult than converting a color photo to black and white :)