Jpeg encoding performance on agx nx jp5.1.1

Autonomous Machines Jetson & Embedded Systems Jetson AGX Xavier
,
1

Hi,
Currently , I that jpeg encoding takes up a lot of memory when adapting the system of jp5.1.1.

img_v2_04a42660-7f02-46f2-a5b1-0dc9c4d71ach_MIDDLE
img_v2_04a42660-7f02-46f2-a5b1-0dc9c4d71ach_MIDDLE1440×1080 234 KB

img_v2_fedc13f7-3efd-4328-bf46-80ed129d7b1h_MIDDLE
img_v2_fedc13f7-3efd-4328-bf46-80ed129d7b1h_MIDDLE1440×1080 203 KB

The way I test it is to start three jetson_multimedia_api/samples/09_argus_camera_jpeg of three camera.
The input image size for jpeg encoding is 1080P@10 YUV420.
I want to know what is causing this problem. Why everything was fine when I was in JP4.6.
regards

2

Can someone help me look into this issue?

4

Hi,
There is a known memory leak in Jetpack 5.1.1 when doing JPEG encoding. Please upgrade to 5.1.2 and give it a try.

5

Is there a quick upgrade command?

7

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