NMR is a media technology company that needed a video management application. There was a massive amount of data to scan and loads of videos to analyze, all of which we converted from a rather useless database into a handy application for business departments to mine, analyse, and acquire insight from.
React, Rx.JS, Ant Design, AWS
ReCAP is the web application that manages the administration, storage, and UI of a solution. Ulam Labs performed frontend development on a web application based on the designs provided, and the frontend development covered various aspects, including:
◦ Facial recognition.
◦ Product placement (for example logos).
◦ Audio recognition.
◦ Video streams on a real-time basis.
The project came with significant challenges:
◦ A web application that analyses loads of videos and photos with ease.
◦ A lightweight video player that’s easily navigable between shots, faces and logos.
◦ Essential metadata hidden in the files presented in a user-friendly, simple way.
We provided the client with a sophisticated, user-friendly frontend application that quietly processes massive amounts of video data. The resulting data can be used in all sorts of exciting and commercially valuable ways by various business departments in their everyday work.
The user interface and main parts of the system were very carefully planned and designed to create a user-friendly dashboard. The relevant data was grouped to keep all the information on one screen: shots, logos, faces and audio.
We worked hard to create a lightweight player. Users can easily navigate between different shots, faces, and logos across the video, and changing a shot has a direct effect on the player.
The improvements we made to the uploader itself make a huge difference. Now the user uploads the file easily, then it’s processed and nicely presented.
An important aspect of the project was the optimized web browser, which had to ensure fast, intuitive research across different categories.
The directory explorer feature lets users create different catalogues and various ways to categorize metadata.