At VISUA we have built a Visual Classification tool that focuses on extracting the most relevant signals from media, so that you can forget about the noise and focus only on what really matters. Specifically built for the needs of platforms and specialist providers, the technology makes it easier for you to derive meaningful insights for your clients that adds incremental value to your platform. And because it works in perfect synergy with our core logo detection module, brand specific and cross-brand analysis is also possible.
Forget unrelated and unconnected labels that make no sense. VISUA’s Object and Scene technology delivers fully categorised and contextualised data thanks to our unique, massive, and deep hierarchical tagging library. VISUA’s technology extracts signals that unlock previously impossible levels of classification, allowing you to deliver unprecedented insights in your platform.
Classify common objects (dog, car, tennis racket, paper cup) at scale. Return multiple object labels via API with supporting data including confidence level and label hierarchy.
Classify any scene (sandy beach to mountain stream, subway train to a busy restaurant) at scale. Return multiple scene labels via API with supporting data including confidence level and label hierarchy.
Identify objects carrying any targeted brand or mark.
Custom or unusual objects and scenes can be specifically trained to meet your use-case and requirements.
The Objects & Scene module can return multiple labels from a library of thousands of pre-trained object and scene classifications for fast detection. So no additional data or lengthy training is necessary.
Object classifications are organised as a hierarchical tree of labels and highlights their parent types within a wider taxonomy of the object and scene. This gives greater flexibility to analyse data at macro or micro levels of granularity and can be customised to your application needs. E.g. Mammal > K9 > Sausage Dog, Labrador, etc..
Objects and Scenes recognition leverages curated semantic metadata to define connections between Objects and Scenes which are semantically related. These connections are sorted from the most concrete to the most abstract and are available for each Object or Scene recognized.
This API can be used in conjunction with brand and mark detection (logo-centric) or used independently depending on your use-case and requirements.
Objects and scene classification can be applied as standard to all popular formats of images and videos at scale. Lesser known/proprietary formats can also be supported as required.
Out team will focus on understanding what you need to achieve and demonstrate the benefits of our technology, such as: