NVlabs/FAN
Fully Attentional Network is a general-purpose vision transformer backbone achieving state-of-the-art results on ImageNet and domain generalization benchmarks.
Not currently ranked — collecting fresh signals.
star history
This repository provides the official PyTorch implementation of FAN, a vision transformer architecture designed to improve self-attention mechanisms for visual recognition tasks. It includes training and evaluation code along with pretrained models for backbone use in image classification, object detection, and semantic segmentation across standard benchmarks like ImageNet and Cityscapes.
Frequently asked
- What is NVlabs/FAN?
- Fully Attentional Network is a general-purpose vision transformer backbone achieving state-of-the-art results on ImageNet and domain generalization benchmarks.
- Is FAN open source?
- Yes — NVlabs/FAN is an open-source project tracked on heatdrop.
- What language is FAN written in?
- NVlabs/FAN is primarily written in Python.
- How popular is FAN?
- NVlabs/FAN has 484 stars on GitHub.
- Where can I find FAN?
- NVlabs/FAN is on GitHub at https://github.com/NVlabs/FAN.