Compare tf lite with SOD library when using embedded system

SOD (SOD - An Embedded, Modern Computer Vision and Machine Learning Library) is an embedded library. Is there any pros/cons compared with TFLite when we want to deploy the models onto ARM embedded system to do video object detection?

Appreciate any help!

SOD (SOD - An Embedded, Modern Computer Vision and Machine Learning Library) is an embedded library. Is there any pros/cons compared with TFLite when we want to deploy the models onto ARM embedded system to do video object detection?

Appreciate any help!