Metadata Enrichment

The application leverages Computer Vision and Neural Networks to identify objects, scenes, people, locations etc. within an image or a video feed. Additionally, the application uses NLP to convert speech-to-text to identify keywords and generate a summary. The application can analyze videos in real time or offline.



Content discovery, categorization and recommendation is a challenge in the face of ever growing volume of content. Tagging images and videos with quality metadata mitigates the issue somewhat, but not eliminate the problem entirely. One of the issues is the fact that metadata  is tagged manually, an activity that is  tedious, time-consuming and error-prone,especially given the volume and pace that the assets are generated.


By leveraging the capabilities of the application, content owners can automatically tag images and videos with extensive metadata. The additional metadata not only helps content discovery, but does so within the context of the video/image. The metadata can be leveraged to categorize content automatically. Keywords and summary can be used to boost SEO ranking and content recommendation. Using NLP, speech based search can be supported as well.


Use cases

Content Discovery






Online Learning

Social Media Content Creators