AILab-CVC / YOLO-World

[CVPR 2024] Real-Time Open-Vocabulary Object Detection
https://www.yoloworld.cc
GNU General Public License v3.0
4.64k stars 449 forks source link

baseline model (no text fusion) #486

Open taltlusty opened 2 months ago

taltlusty commented 2 months ago

Hello, Is there a config file for the base model that was used to explore the results with no text fusion, as presented in table 4 in the paper? Thanks

wondervictor commented 1 month ago

Hi @taltlusty, we haven't released the checkpoints for ablation experiments. We mainly release pre-trained models for practical purposes.