Real-time Transformer-based Open-Vocabulary Detection with Efficient Fusion Head
Paper • 2403.06892 • Published • 2
OMDet-Turbo (Swin-T) real-time open-vocabulary detector weights mirrored for
LibreYOLO's LibreOpenVocab tier.
Mirrored from omlab/omdet-turbo-swin-tiny-hf. The model weights are published by Om AI Lab under the Apache License 2.0.
Paper: https://arxiv.org/abs/2403.06892
No learned parameters were changed. This repository preserves the Hugging Face
transformers snapshot files needed by LibreYOLO's open-vocabulary detector
wrapper and replaces the model card with LibreYOLO-specific loading notes.
from libreyolo import LibreOpenVocab
model = LibreOpenVocab("omdet-turbo")
model.set_classes(["person", "dog", "car"])
results = model.predict("image.jpg")
Base model
omlab/omdet-turbo-swin-tiny-hf