/spark-augmentation

A robust learning scheme by proposing a novel augmentation algorithm that scales to emulate a large set of image degradation and occlusion policies

Primary LanguageJupyter NotebookMIT LicenseMIT

spark-augmentation

list of augmentations

A robust learning scheme by proposing a novel augmentation algorithm that scales to emulate a large set of image degradation and occlusion policies.

Paper: Residual Net for Car Detection with spatial transformation (2020)

Code: spark_augmentation.ipynb & spark_augmentation_erase.ipynb