Features and Pricing

Our API service analyzes a high volume of images, harvesting demographic and other features pertaining to people's faces. Our customers will benefit seamlessly from our ongoing algorithmic enhancements through our API. You send us images and we send you JSON objects with corresponding results. Our massively scalable ImageSleuth system runs on Amazon web services. We take pride in helping our customers find great solutions. We thrive on custom jobs creating new features. Processing millions of images is our specialty.

Why to choose us

Our ImageSleuth API is applicable across multiple vertical markets, including retail, marketing, financial, search and security. We provide insights about who is coming to your store or mall, and who is purchasing. ImageSleuth features may also be used to index and rapidly search your media to find people who meet demographic and other criteria. For example, find Asian men in their 30s with glasses. Security systems, which produce significantly more data than can be reviewed by human operators, would greatly benefit from our ability to rapidly index that data, making search more timely and effective.

ImageSleuth employs a new technology – HYDRA – which merges the latest advancements in computer vision and machine learning. The complexity of HYDRA is completely hidden behind the API, making ImageSleuth extremely easy to use. HYDRA recovers demographic features like age, race and gender. Our technology recovers other details that can be useful to search on, like hair and eye color, the existence of facial hair, eyewear, a hat. We also provide insight into the subject’s mood (is she smiling?), as well as the pose of the head and whether the subject is looking at the camera, and more. Hair and eye color are helpful in retail to establish color schemes for clothing and other merchandise. The focus of the subjects’ eyes is useful in out of home networks, like in gyms and airports, where we want to capture information only from those who are looking at the ads (assuming the camera is close to the screen showing the ad). Security applications could tap into vague details like those collected from witness accounts, e.g. whether the subject had a beard and/or was wearing a hat.

Image Sleuth is a cloud-based distributed architecture that processes 3 faces per second per core and scales linearly with the number of cores. ImageSleuth is currently deployed as a web service on AWS. Our instances are load balanced, i.e. they are designed to distribute their processing of images optimally across existing instances. If the loads require it, our AWS cluster can launch additional instances, guaranteeing that all our customers will experience excellent performance on their datasets, no matter how large they might be.

ImageSleuth returns your data in Big Data friendly JSON format that provide outcomes and beliefs in those outcomes: e.g. how certain are you that the subject is Hispanic, male, and in his 20’s? Instead of a one size fits all results, providing beliefs enables our customers to use their favorite Big Data platforms to decide how and whether they will choose to employ the outcomes. For example, one customer may only be interested in results with a high degree of certainty, and will only accept outcomes with high degrees of belief, while another may tolerate more uncertainty. This allows end users to make individual informed decisions and customize reports for their use.



Process 300 Images


Process 10,000 Images


Process 100,000 Images


Process 1,000,000 Images


Process As many images as you need

Real Use Cases