AI NewsWaveTwelveLabs Releases Pegasus 1.6, a Video Understanding Model for Robot Learning
Key points
- Refines first-person video into robot learning data
- Vision-language model Pegasus 1.6 released
- Business expands from media to physical AI
TwelveLabs said on the 7th that it has released Pegasus 1.6, a vision-language model (VLM) designed to understand first-person video shot from a human's point of view, known as egocentric data.
The model automatically identifies actions, objects and task sequences in video and refines them into robot learning data. IT Chosun reported that it can reduce the time and labor needed to process data.
AI Times explained that TwelveLabs is extending the video understanding technology it has applied in entertainment, sports and the public sector to physical AI fields such as robotics.