How to run LPR with .etlt model

• Hardware Platform - GPU
• DeepStream Version-5.1
• TensorRT Version-7.2
• NVIDIA GPU Driver Version (valid for GPU only)-460

By referring [Creating a Real-Time License Plate Detection and Recognition App | NVIDIA Developer Blog] able to run the deepstream app using LPD-LPR pretrained ‘etlt’ model of US.

For running the same on another devices again need to manually convert the ‘.etlt’ model to ‘.engine’ by using ‘tlt-converter’ commands on the particular devices.

Is it possible to run the same app using direct ‘.etlt’ model? Like the other sample deepstream app which create the engine file by itself while running.

Current deepstream does not support ONNX generated etlt model parsing. Please use tlt-converter tool to generate engine files.

okay…thanks for the clarification.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.