LIVE LLAVA with VILA1.5-3b, no events alert after adding content in Filters and Tags

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin NX
1

Hi all,
We are trying to use Live LLAVA demo on Orin NX16G+ Jetpack 6, we can run the Live LLAVA successfully and the response of prompt is correct.

When we set the keywords on Filters and Tags of events box, no events were detected even though there are already objects related to the filter on the screen.

Does anyone know how to enable the event filter function on Live LLAVA demo?

3

Hi,

Could you share the way you set keywords so we can give it a check?
Thanks.

4

Hi Aasta,

Step1:
jetson-containers run $(autotag nano_llm)
python3 -m nano_llm.agents.video_query --api=mlc
–model Efficient-Large-Model/VILA1.5-3b
–max-context-len 256
–max-new-tokens 32
–video-input /dev/video0
–video-output webrtc://@:8554/output

Step2:
Chromium Browser to open link https://127.0.0.1:8050

Step3:
See the live streaming from web camera, change the prompt can get the correct response.

Step4:
In the Events field, add keyword “detected people” in Tags and add keyword “people, person, man, women” in Filters. But no any event alert was added in the event form even though the response of prompt shows someone is present.

5

Hi.

Thanks for the details.
Will provide more info to you later.

7

Hi,

Please check if the below topic helps with your use case:

Thanks.

9

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