World leading visual intelligence solutions
Pimloc's deep learning platform allows public and private organisations to automatically detect, classify and anonymise content in large-scale image and video collections or streams.
Deep AI expertise
World-class machine learning pedigree. Deep learning proprietary neural networks training using millions of images.
World-leading models for popular and specialist still and video content. - Faces - Objects - Scenes
Image discovery, search, ranking, specialist user training of long-tail content. “Finding a needle in the haystack”
Secure & flexible
Platform scalable across local, edge and cloud processing to optimise solutions for data volume, speed of upload/processing and data security
How the platform works
The Pimloc platform has been trained by showing it millions of images across a wide range of content. It has learned how to recognise and classify faces, objects and even scenes in images it has never seen before.
The platform is accessible through Pimloc’s Search and Discovery App, by using the Pimloc APIs or through a range of tailored Pimloc’s Solutions. The platform can be run in the cloud, remotely, or stand-alone. It can recognise and classify general or specialist content with world-class accuracy and speed. The Pimloc App allows users to use their own specialist knowledge and experience to increase the performance and capabilities of the system.
The Platform can automatically search for content that matches a built-in catalogue of over 20,000 well-known objects and scenes. Searches for unknown objects and scenes is also straightforward, using reference images provided by the user or obtained “on-the-fly” from the Internet. The platform learns image by image; creating new classifiers dynamically as users search through your content.
Pimloc Platform Architecture
Increase the value of imagery and image analytics with higher productivity and lower costs
Pimloc platform features
Pimloc puts the power of vision-based machine learning (AI) directly into the hands of your internal teams or online service users. The platform locates, identifies and learns frame by frame, all within the security of your own business systems – with no need for any manual tagging.
Image Library Search
Pimloc has developed a solution for image libraries to make their large and diverse collections instantly discoverable. Pimloc is working with a range of archives, including leading art and heritage institutions.Find Out More
Brand Tracking in Video
Pimloc has developed an automated system for tracking brands (and products) within video footage. This solution is currently being used to track the quality and length of on-screen brand appearances within sports footage for a leading sports media company.Find Out More
Remote Image Processing
Pimloc has created a remote processing solution, that can run fully offline, to provide on-site image and video processing services - for situations where upload speed, search latency and data privacy are a critical challenge.Find Out More
Tracking human activity
Pimloc is developing automatic solutions that can accurately track human pose and activity from 2D and 3D video; focussed on a range of use-cases and environments.
Auto Publishing (beta)
Pimloc has developed a publishing service that analyses articles as you write them so it can automatically suggests relevant images to be included from networked drives and image library subscriptions.Find Out More