Hello, and thank you for sharing your open-source project.
I’m currently testing the code on a Jetson Orin NX device in MAXN mode. The issue is that even when I run only the Depth component (excluding VINS), the GPU usage exceeds 90%. In the paper, it says the GPU usage is around 40%, and VINS supposedly uses about 5%. However, even when running only VINS, I’m seeing a GPU usage of about 40%.
I’m wondering if this discrepancy is due to differences in the default TRT model provided, or if there might be some other underlying issue.
Hello, and thank you for sharing your open-source project.
I’m currently testing the code on a Jetson Orin NX device in MAXN mode. The issue is that even when I run only the Depth component (excluding VINS), the GPU usage exceeds 90%. In the paper, it says the GPU usage is around 40%, and VINS supposedly uses about 5%. However, even when running only VINS, I’m seeing a GPU usage of about 40%.
I’m wondering if this discrepancy is due to differences in the default TRT model provided, or if there might be some other underlying issue.