What's the biggest unused potential in DAM?

The way forward
© Rawpixel - Fotolia.com #88461969

The simple answer: images. Although the usage of images in DAM systems is higher than ever, most DAM systems are not capable to provide efficient search and management tools for visual assets. This blog post answers the why and the how to solve this.

If images were only text documents

Today, almost all DAM vendors are able to provide an advanced text search and document analysis. Documents like PDFs or Word files can be added to the catalogue without further user interaction.

The DAM software automatically analyses the document, extracts and index keywords to make it searchable. Users do not need to add metadata to that document to make it findable at a later time. Metadata like tags or category may be added, but this is not required. Instead, the information that is already present in the document itself is used (the keywords, headings etc.).

It is obvious why using the information stored within a document is clever:

  1. It enables a fast and efficient cataloguing process, since the user uploads the document and it is searchable instantly.
  2. The text within the document describes best what it is about. No metadata can be as complete as the text itself.

Why metadata can't be the solution

While text documents can be analyzed automatically, visual assets have to be tagged manually in order to make them findable. Visual assets are still represented, managed and searched by metadata. Therefore, the user still does not search for images, instead the user searches for metadata. And if no sufficient metadata is present, no suitable image can be found.

Manual tagging is time-consuming, inconsistent, incomplete and error-prone. This leads to a cumbersome workflow and causes unnecessary expense.

Images contain information about themselves

Instead of using and relying on the metadata of visual assets it's much more reasonable to use the visual information stored within the image pixels like we already do with text documents. All you need to do is to extract that information.

That's where we at pixolution come into play. Our product pixolution flow automatically transforms the so far unused image pixel data into useable information to work with. This way users are enabled to do an image search by entering visual search parameters (for example sample images or color schemes). Furthermore, users avoid to have duplicates in their catalogues and thus prevent multiple image licensing costs. Optimizing the cataloguing process by getting tag suggestions is just a further work simplification possible with pixolution flow.