DAM Guide – Shape and location recognition in DAM software

Digital asset management software is equipped with a tool for recognizing shapes, which makes it easier to index and search for image and video files.

With other systems, indexing and searching for files is generally done based on metadata. Those metadata can actually include text information that describes the content of the file.

With digital asset management software, indexing and searching for files is no longer solely based on the metadata attached to each file: the software can actually access file content. This is possible thanks to the usage of shape recognition tools. Within its photo library, a company can index and search for files based on criteria or similarities. In concrete terms, when you select an image to be used as a reference by the recognition tool, the DAM software will find all other images with similar shapes. This applies to the shapes of objects, buildings, landscapes, logos, and even faces. In this way, the DAM software is able to detect and display all similar images.

More broadly, shape recognition can sometimes go further than simply recognizing geometric shapes and can extend to recognition of all sorts of items: letters, numbers, bar codes, text structures, color palettes, and even sound patterns.

Thanks to shape recognition for inanimate objects and facial recognition, DAM software can optimize your searches for image and video files by automatically grouping similar files together. Thanks to recognition tools, companies and public-sector entities can save time and boost their productivity.

For example, in Ephoto Dam, all you have to do is click on the thumbnail of a face present in a photo to launch a search for that person and thus find all photos in which they appear.

Shape recognition and facial recognition both rely on artificial intelligence, producing benefits for the user.

First, we have to define what artificial intelligence is: it is “intelligence” given to a machine so that it can take the place of a human being in performing a task. The task in question is one for which that human would have to use their intelligence. In this case, artificial intelligence supports DAM by taking the place of the human eye when comparing images to each other. That means that you no longer need a human to compare images: the machine takes care of it, saving you a lot of time. Indeed, it could take a human being hours and hours to compare images; meanwhile, the machine is able to perform the same task in a matter of minutes, or even a few seconds, depending on the size of the photo library.

When it comes to facial recognition, artificial intelligence always works in three successive stages: first, a face is detected; second, the face is analyzed, resulting in a digital model; third, the recognition algorithm compares the digital model with that of the reference image. Thus, without need for a human eye, facial recognition and machine learning make it possible to automatically recognize faces and sort files according to the people appearing in them.

The principle behind it is roughly the same as the principle behind the recognition of other shapes such as buildings and objects. The artificial intelligence detects a shape, analyzes it, and compares it.

Shape and facial recognition are not the only tools that make it easier to index and search for files: there’s also a feature that will detect the geographic locations where files are from. That way, you can find the geographic locations of images directly in the photo library.

There are several ways this location feature can be used.

First, sometimes GPS coordinates are already attached to the file (included in metadata) and the software can then index using those coordinates. Thus, the software automatically takes GPS coordinates into account.

In other cases, the file will not have any GPS coordinates attached. The user can then enter those coordinates themself when they add the file to the photo library. This can be done by placing a pin directly on a map; alternatively, the user can type in the GPS coordinates.

Once files are indexed by geographic location, they can be found by searching for an address or GPS coordinates, or by searching directly on a map and selecting a pin.

Like with shape recognition, location services can save your organization a lot of time. Indeed, location services make it possible to automate some of your organization’s multimedia data management, particularly when it comes to indexing and searching by where an image was captured.