When you share your carefully-curated content on social media, you want to ensure it reaches as much of your audience as possible. One important part of this process is picking an amazing image to include in that post. That’s why we created the Social Media Image Recommender, an algorithm which looks through your article’s text and images, and tells you which picture you should use when sharing your post on Facebook, Twitter, LinkedIn, or any other social network. As a bonus, it also smart-resizes your image to the correct size for social sharing, keeping the most important elements in the frame and discarding unimportant background, thanks to the magic of deep learning and machine vision.
Want to jump right in and try it out yourself? Check out our Image Recommender Demo! Like it? Sign up and integrate the Image Recommender Algorithm into your digital workflow to process thousands of articles every hour, and supercharge your content publishing pipeline!
When we implemented InceptionNet, a microservice to detect and label objects (features) in photos, we knew it would be helpful. Then, we built out VideoMetadataExtraction, a video pipeline which allows you to run feature-detection algorithms (and others) on an entire video. This allowed for some really powerful activities — like automatically scanning through home security footage to find all the cars of a specific make & model, or stripping out all the nudity-containing scenes of a movie to make a G-rated version.
Add emotion detection to your livestreaming service. Find all the faces in your security tapes. Detect the make and model of every car that passes your shop. Flag nudity in users’ uploaded videos.
These are just a few of the cool things you can do with our powerful Video Metadata Extraction microservice, which allows you to analyze an entire video with our many image data-extraction algorithms. But don’t simpy take our word for it — if you want to see how powerful this tool is, take a peek at our live demo:
Recently, we introduced you to Video Transform, a meta-algorithm which can take any of our image transformations — such as Colorization, Saliency, Style Transfer, or Resolution Enhancement — and apply it to a video instead!
…but if picture is worth a thousand words, a video is worth 24-30,000 words per second (sorry, bad videography humor there). So instead of just telling you about this cool feature, we’d like to show you, with a brand new Video Toolbox demonstration!
Once upon a time, site mappers were arcane scripts which could take hours or days to examine a single website. But, thanks to scalable & interoperable cloud algorithms, it now takes only minutes… and includes a multitude of handy features powered by machine learning: auto-tagging, summarization, page-ranking, and more!
- GetLinks recursively traverses a website of your choice, plotting them on a force-directed graph via D3
- PageRank examines the pages to create an ordered list akin to Google’s PageRank Algorithm
- Url2Text grabs the text from each page, allowing Summarizer to extract topic sentences while AutoTag generates keywords