OWL-ViT
Open-vocabulary object detection model by Google using vision transformers.
About
OWL-ViT, Vision Transformer for Open-World Localization from Google, performs open-vocabulary object detection by accepting free-text queries rather than a fixed label set. It transfers image-text pretraining in the style of CLIP to detection without task-specific training data, so it can localize objects described by arbitrary text. It is distributed within Google's Scenic research codebase for attention-based vision models. Released under the Apache 2.0 license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Intermediate (3/5)
- License
- Apache-2.0
- Minimum VRAM
- 6 GB
- Added
- Apr 3, 2026
Related Tools
Foundation model for monocular depth estimation by TikTok.
Improved vision-language model by Google using sigmoid loss for contrastive learning.
OpenMMLab detection toolbox with 300+ pre-trained models and 80+ algorithms.
State-of-the-art real-time object detection supporting YOLOv5 through v11.
Robust multi-object tracking combining motion and appearance cues.
Additive angular margin loss for deep face recognition.