ENEAS

Text-promptable instance tracking and segmentation across images and video

speridlabs.com/research/eneas

Community-submitted · content unverified. This organization has not verified ownership or the information in this profile.

About

ENEAS (Embedding-guided Neural Ensemble for Adaptive Segmentation) is Speridlabs' unified, text-promptable method for instance tracking and semantic discovery. Given a natural-language prompt and a set of frames, ordered video or an unordered image collection, it returns masks that follow one specific instance through occlusion and scale changes, or finds every instance a query names, while telling real objects apart from statues, paintings or reflections. Code and a demo Space are public.

Added via
web
Ownership
unclaimed

Where to go

  • Homepagesperidlabs.com/research/eneas
  • GitHubgithub.com/speridlabs/eneas
  • Homepagesperidlabs.com/research/eneas
  • GitHubgithub.com/speridlabs/eneas

Product details

Product kind
Open Source
Open source
Yes

Social profiles

Links