search

naver / shine

[CVPR'24 Highlight] SHiNe: Semantic Hierarchy Nexus for Open-vocabulary Object Detection
Other
42 stars 2 forks source link

How reproduce the results of the paper! #1

Closed zhongzee closed 3 months ago

zhongzee commented 4 months ago

Hello, thank you for your excellent work. I have only seen the train script so far. Which script should I run to verify or reproduce the results of your paper?

OatmealLiu commented 4 months ago

Hi! Thank you for your interest in our paper! There is no training needed/involved by SHiNe. In the README.md file, we provided an example of running scripts to reproduce the results in the paper.

Specifically, the evaluation scripts for OvOD task are placed in shine/scripts_local's CoDet, Detic, and VLDet, respectively. The ones for classification task are placed in shine/scripts_local/classification. If you wanna run it with slurm, you can simply switch scripts_local to scripts_slurm.

The train scripts under the root folder are the entry points (or main functions) these scripts will call out.

Best wishes!