Difference in Facial Recognition Accuracy between Jetson Nano and Jetson TX2

@snadeem.bee17seecs is using a multilayer perceptron (MLP) on Google Colab using the same versions of Python, OpenCV, Tensorflow (2.3.1), and Numpy. There is a 1% difference in accuracy in facial recognition between the Jetson Nano (98%) and Jetson TX2 (99%) even though the same model and test video are used. What could be the cause of differences in accuracy?

@snadeem.bee17seecs - I moved your Deep Learning on Jetson Devices … help needed! - AI & Data Science / Computer Vision & Image Processing - NVIDIA Developer Forums here.

Hi, @snadeem.bee17seecs

Is this still an issue for you?
Since we have several new software releases, please help to test this on the latest JetPack if this is still a concern.


This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.