Image recognition tools for digital asset management software
Artificial intelligence is a hot topic right now, and not just in digital asset management but everywhere. Some people are positive about the future of AI and then there are those that think robots are going to take over the world. However, for DAM systems, we’re just talking about AI as image recognition and how it can help tag your photos. So, we don’t have to worry about robots taking over the world, at least not yet.
Nonetheless, there are those who argue for and against AI in DAM as well. They say it won’t save the admin any time, while others say it will save admins a ton of time.
Regardless of what your position is, we’d like to educate you on the image recognition tools that are available for DAM right now. Just keep in mind that the tools work through an API, which means that not all of them will work exactly the way we describe them below. The tools can be customized to fit your needs. You’ll have to reach out to your individual vendor to see what tools they offer.
Did we miss something? Let us know in the comments or tweet us @digitalassetmgt.
Image recognition tools
“A service that makes it easy to add image analysis to your applications.”
Amazon Rekognition can identify objects, scenes, people, celebrities, and detect inappropriate content in images and videos. Additionally, there’s facial analysis and recognition. Facial analysis means that the technology analyzes the attributes of faces to determine characteristics such as emotion, age range, glasses, facial hair, etc. There’s also textual analysis which can detect and recognize text from images such as product names, street names, captions, and license plates. It detects all objects with a confidence score. This means the AI will let you know how sure it is that what it’s detecting is actually what it’s detecting. This technology doesn’t require any machine learning expertise to use. You only pay for the number of images or minutes of video that you actually analyze and the facial data you store for facial recognition. There are no minimum fees or up-front commitments.
“Artificial intelligence with a vision.”
Clarifai automatically tags all of your images. Clarifai can recognize over 11,000 general concepts (outdoors, indoors, person, etc.) and can be trained to tag objects more specifically as well (e.g., naming your business headquarters rather than just tagging it “building”). You can also automate moderation of content. It can flag nudity, illegal substances, weapons, or you can customize what you would like it to detect, such as off-brand content. Furthermore, Clarifai works in 20 different languages. Clarifai has three pricing tiers: free, standard, and enterprise.
“High quality understanding of images within seconds.”
CloudSight automatically tags all of your images and can tag product-specific details such as brand, style, type, and more. Additionally, it can flag images for inappropriate content. The pricing ranges from $99-$899 per month and is based on the number of image requests per month. They offer a free trial to get you started. Try it out here.
“Powering the image search of tomorrow.”
EyeEm automatically tags all your images with concrete objects and concepts. Furthermore, it generates captions for each image by combining tags with semantics. Plus, it can determine the aesthetic quality of your images and allows for personalized ranking based on a specific visual requirement. EyeEm is also a repository of stock photos, so there is pricing information about their stock photos but not for their image recognition tool. See a live demo.
“Derive insight from images with our powerful Cloud Vision API.”
Cloud Vision automatically analyzes images and provides thousands of relevant tags such as sailboat, Eiffel Tower, lion, etc. Likewise, Cloud Vision can detect faces and text within images. It’s used to add metadata to your image catalog, perform image sentiment analysis, and moderate offensive content. Cloud Vision uses Google Safe Search to moderate inappropriate content such as adult material or violence. Accuracy is improved over time as new concepts are introduced to the software. Not only does Cloud Vision detect text in images, but it can also detect the language text is in and offers support for multiple languages. Other types of recognition include logo detection, landmark detection, image attribute detection, and similar image detection. Google Cloud Vision prices their service based on the number of images processed per month and the type of processing used. Prices range from free to $3.50 per image. See it in action.
“Build your apps on top of an advanced image tagging technology!”
Imagga automatically tags your images and becomes smarter as time goes on because the algorithms learn your set of metadata and adapt to your needs. Imagga has thousands of predefined tags, but it can also be trained to suit your needs. It allows you to define tags and then the software learns to recognize and suggest them. Pricing starts at $14 a month for 12,000 images and goes up from there. You can start by signing up for a free trial. See a live demo here.
Have you used any of these tools in your DAM system? Let us know if it’s saving you time tagging your assets in the comments! And as always, reach out with any questions you have.