search

Tobias-Fischer / rt_gene

RT-GENE: Real-Time Eye Gaze and Blink Estimation in Natural Environments
http://www.imperial.ac.uk/personal-robotics
Other
365 stars 68 forks source link

Use trained model for inference #38

Closed sahilrider closed 4 years ago

sahilrider commented 4 years ago

I have trained the model as per the rt_gene_training folder. I have saved the final models for each folds. I want to use these models to test on new images. How to do it? Which model to use?

I don't want to go through the whole ros pipeline. Thanks

Tobias-Fischer commented 4 years ago

Hi, Please see the comments in #37; in particular you can now simply use https://github.com/Tobias-Fischer/rt_gene/blob/modularised/rt_gene/scripts/estimate_gaze_standalone.py (don't forget to change branches). The script allows you to specify an image path as well as which model files to use.

Best, Tobias

Tobias-Fischer commented 4 years ago

See also https://github.com/Tobias-Fischer/rt_gene/blob/modularised/rt_gene_standalone/README.md