lucidrains/slot-attention
PyTorch implementation of Google AI's Slot Attention, an attention mechanism for object-centric learning and clustering.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch implementation of Slot Attention, a neural network architecture that uses a differentiable attention mechanism to decompose data into distinct slots representing objects or entities. The implementation includes the core SlotAttention module with configurable iteration counts, and extends to MultiHeadSlotAttention and AdaptiveSlotWrapper for generating soft slot masks. The model learns to segment and represent multiple objects in a scene without explicit supervision on object boundaries.
Frequently asked
- What is lucidrains/slot-attention?
- PyTorch implementation of Google AI's Slot Attention, an attention mechanism for object-centric learning and clustering.
- Is slot-attention open source?
- Yes — lucidrains/slot-attention is open source, released under the MIT license.
- What language is slot-attention written in?
- lucidrains/slot-attention is primarily written in Python.
- How popular is slot-attention?
- lucidrains/slot-attention has 491 stars on GitHub.
- Where can I find slot-attention?
- lucidrains/slot-attention is on GitHub at https://github.com/lucidrains/slot-attention.