Best approach to detect 2 classes in images on a Nano 2GB?

wiasmitinow · September 9, 2021, 11:23am

Hi All

We know 2GB is not a lot. So how to use them best? I want to detect 2 classes: boxes and covers as shown in image below. It works great on my PC but we’re running out of memory on the Nano 2GB using Pytorch and YoloV5s.
What would you recommend?

AastaLLL · September 15, 2021, 3:01am

Hi,

You can check our jetson_inference sample.
The sample converts a detector into a TensorRT engine which requires less memory.

Thanks.

Topic		Replies	Views
Using 256GB MicroSd Card with Jetson Nano Jetson Nano jetson-inference	2	1373	October 18, 2021
How To Use Detection Model From Other Sources? Jetson Nano jetson-inference , nano2gb	3	527	October 15, 2021
Using Object Recognition on Jetson Nano 2GB for 2D Applications Jetson Nano jetson-inference , nano2gb	2	702	October 15, 2021
Improve inference performances yolov5 Jetson Nano yolo , nano2gb	4	1542	June 2, 2022
Jetson Nano faster for object recognition with GPU Jetson Nano jetson-inference , nano2gb	5	987	December 15, 2021
Inference OD Memory Consumption Jetson Nano jetson-inference	5	539	March 10, 2022
Jetson Nano GAN Generative Adversarial Network Jetson Nano jetson-inference , nano2gb	2	1417	October 15, 2021
Jetson nano + custom data + training and optimisation TensorRT tensorrt , yolo	2	488	September 21, 2021
Jetson Nano 2GB Tensorflow Object Detection API/TensorRT, Ran out of memory Jetson Nano tensorflow , nano2gb	2	488	October 15, 2021
Jetson Nano 2GB vs. 4GB Jetson Nano camera , nano2gb	4	3149	October 15, 2021

Best approach to detect 2 classes in images on a Nano 2GB?

Related topics