Our Images and video recognition models simplifies the addition of image and video analysis capabilities to applications. Simply provide an image or a video to our API and the service can identify objects, people, text, scenes and activities, as well as detect any inappropriate content. Our service also offers extremely accurate functionality of facial analysis and facial recognition. You can detect, analyze and compare faces for different use cases of user verification, cataloging, people counting and public security.
Our service continuously learns from new data to expand the ability to recognize objects, scenes and activities and improve accuracy. With our recognition solution, yours applications automatically take advantage of these continuous improvements.
Our alghoritms offers constant response times regardless of the volume of requests made. The application latency remains constant, even if the volume reaches tens of millions of requests.
Benefit implementing our solutions
- Simple integration of powerful image and video analysis capabilities in an app
- Continuous learning
- Fully managed service
- Batch analysis in real time
- Usable anywhere
|IMMEDIATE RESPONSE FOR PUBLIC SECURITY
Our AI video elaboration models let you create applications with real-time recognition of people of interest from live streaming of cameras based on a private metadata database related to faces. Cathedral's alghoritms also allows you to quickly and easily review hours of video footage looking for people of interest, monitor their movements and detect their activities.
|VIDEO LIBRARY AVAILABLE FOR RESEARCH
Our video solutions automatically generates metadata from uploaded videos, allowing you to create a search index for celebrity names and when they appear. You can keep your index up to date using yours or our triggre functions to automatically add new video labels to the search index when a new video is uploaded to a dataset. You can then use this index with your DB to quickly find video content.
|NOT SECURE VIDEO DETECTION
Cathedral's ML video models allows organizations that manage user-generated content, such as dating or social media apps, to automatically detect explicit or vulgar content in videos and create custom rules to define what is appropriate for culture and demographic data of users.
|IMAGES LIBRARY AVAILABLE FOR RESEARCH
Our images solutions makes images available for searching by allowing you to find objects and scenes within them. You can create a trigger that automatically adds labels to new images detected directly in an dataset index when a new image is loaded into DB.
Our alghoritms automatically detects images with inappropriate content using the Amazon AWS Image Moderation API. This API returns a reliability score based on a detailed set of categories, which allow you to create your own rules on what is appropriate according to the user's culture and demographics.
|USER VERIFICATION BASED ON THE FACE
With our images ML models, your applications can confirm your identity by comparing your live image with a reference image.
Our models are able to detect emotions like joy, sadness or surprise from facial images. Here in Cathedral we can analyze live images and send emotions related to DB for periodic reporting on trends for each store.
We can simplifies the search for similar faces in your image collection by storing facial metadata with our solutions. Later you can use our API function to return high security matches. A collection of faces is an index of faces you own and manage.
|VERIFICATION OF THE CAR'S PLATES
With our powerful ML images solutions, your applications can identify vehicles that pass in front of a camera by recognizing the license plate