Video-to-text analysis
Analyzes video and generates context-aware text such as summaries, tags, timestamps, or structured output.
Pegasus 1.6 by TwelveLabs
Convert first-person video into robot-training data
The overview
What is Pegasus 1.6 by TwelveLabs?
Pegasus 1.6 is a TwelveLabs video model for converting egocentric, or first-person, footage into data for robot training. The launch focuses on footage captured from a person’s point of view and its use in robotics training. TwelveLabs describes its broader platform as supporting video search, analysis, and text generation.
Around the internet
What people are saying
Public experiences, questions, and discussions.
No discussions collected yet.
Inside the toolkit
Key features
First-person video understanding
Supports analysis of egocentric footage captured from the viewpoint of a person or device.
Improved people and object recognition
Recognizes people, characters, and objects in video with improved identification.
Native image analysis
Analyzes still images through the same API and prompts used for video.
Where it fits
Use cases
- Label first-person recordings for roboticsSegment and label body-camera, robot-camera, or teleoperation footage for robotics and embodied-AI workflows.
- Catalog and review media or security footageIdentify people and objects across video for media cataloging or security review.
- Analyze still images and video in one integrationUse the same API for mixed image and video content, such as product photos with product videos.
Keep exploring
