Keras implement of Stacked Hourglass. The model is for keypoint detection, and can be migrated to other tasks.
If you're using tf2.0 or higher, replace keras with tf.keras at import.
Way to build a model is exampled in StackedHourglass.py.
Full train-eval-test pipeline is exampled in test.py.
After inference OKSEvaluator.py can be used to evaluate average OKS of the result.
-
Notifications
You must be signed in to change notification settings - Fork 1
feiyangsuo/Stacked-Hourglass-for-keypoint-detection
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Keras implement of Stacked Hourglass. The model is for keypoint detection, and can be migrated to other tasks.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published