Is it possible to run Faster-RCNN in TensorRT with Jetson-TX2 at real-time FPS?

I’m trying to implement Faster-RCNN in jetson-tx2 with tensorRT, but has anyone got the results for the same with Real-time FPS, i mean 30 fps?

moving to TX2 for support coverage.

Duplicated with [url]https://devtalk.nvidia.com/default/topic/1044537/can-we-get-real-time-fps-30fps-with-faster-rcnn-caffemodel-tensorrt-and-jetson-tx2-/[/url]