This repository has been archived by the owner on Dec 8, 2021. It is now read-only.

Make DeepLabV3+MobileNetV3 Backbone runnable with trt on NVIDIA Xavier #7

Open

alexanderwendt opened this issue Mar 11, 2021 · 0 comments

Assignees

Labels

Contributor

alexanderwendt commented Mar 11, 2021

We need to get Bernhard's DeepLabV3+MobileNetV3 runnable with trt on NVIDIA Xavier to be able to utilize the hardware better.

For native trt, the model has to be converted from H5 to ONNX

The execution environment shall be the same as for object detection, using the latency.csv and performance.csv as exchange formats

alexanderwendt added the enhancement label

alexanderwendt assigned themozel

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.