VIDEO RESEARCH
Discover how Visual AI can help you with Video Analytics
Video Analytics is at the forefront of application of artificial intelligence for analytics.
Our Video Analytics engine is based on the same proven, highly scalable, deep learning technology that powers our image analytics platform. You may receive your video content from multiple sources such as IOT devices, CCTV or mobile phones or it could be in your own library as curated content.
VIDEO FROM THE EDGE
IOT
CCTV / MOBILE
Live stream
CURATED CONTENT
Library
We analyze videos in all formats
Our comprehensive approach to video analysis allows you to extract various details including Object Detection, Celebrity Identification, Convert to Voice to Text, Content Moderation, sequence identification, vehicles, weapons, nudity and on screen text to name a few.
Object Detection or scene detection refers to the process of analysing a video to assign labels to objects based on its visual content. Object detection is ideal for customers, who want to search and organize large image libraries.
Our object detection algorithm can identify 1000’s of objects including but not limited to: People and events such as wedding, birthday, bride, guitarist, etc, Food and Drink such as pizza, apple, wine, cake, sandwich,etc., similar entities in Nature and Outdoor, Animals, Home and Garden, Transportation, Electronics, etc.
Face analysis is the process of identifying a face and extracting relevant face attributes from it. We can detect 100 faces in an image or video frame. You can locate face attributes such as whether the face is smiling, eye are open or show emotions.
For each face detected we can also tag the face with information such as gender, smile, emotions, eyeglasses, eyes open, mouth open, moustache, beard, pose, quality, face landmarks, etc.
Vehicle detection is the process of identifying a vehicle and assign category labels to it. We can detect 100 vehicles in an image or video frame and the vehicles such as cars, trucks, buses, two wheelers, cyclists, pedestrians can be identified.
From a video sequence the number of vehicles passing through a point, the speed of the vehicles, direction of the vehicles – up or down are identified. Vehicle detection finds application in gate control and traffic enumeration.
Historically customers have to work with transcription providers to convert an audio to text. We use deep learning method called automatic speech recognition that quickly converts audio to text.
Our transcription models can be used to transcribe customer service calls, automate subtitling, and generate metadata for media assets to create a fully searchable archive.