-
**root@3b9d4277941b:/opt/savant/src# python module/run.py**
INFO insight::savant::config::module_config > Configure module...
WARN insight::savant::config::json_resolver > JSON loads fail, re…
-
The update from 1.40.18 to 1.44.2 (commit 959cd839) broke the ability to make VPN connections, as dns-data doesn't exist. Would it be possible to somehow detect the host version of NetworkManager, rat…
-
Code content:
Code that receives pgie as tensor-meta and draws bbox on osd (later Add align - sgie process). In order to `get_nvds_buf_surface` in the code, the **buffer must be RGBA, so set the sou…
-
Performing the steps at [NVIDIA-Jetson-Nano-Deepstream---Docker-Example](https://github.com/cu-ecen-aeld/yocto-assignments-base/wiki/NVIDIA-Jetson-Nano-Deepstream---Docker-Example) to Build a Yocto Te…
-
I assume the jetson inference library uses the CUDA frame for performing inference.
Could you please give me a sample Gstreamer pipeline that takes CSI camera input and gives CUDA frame output to ap…
-
配置文件如下:
app:
type: video_decoder
enable-perf-measurement: true
perf-measurement-interval-sec: 5
nvurisrcbin:
uri: rtmp://localhost:55555/live/0.mp4
gpu-id: 0
nvvideoconvert:
…
-
您好,有个问题想要请教一下:
1.关于nvinfer,我看您是修改了gstnvinfer.cpp达到了自定义插件,请问这个插件是一定得在nvinfer的时候使用吗?如果我想在流水线中将仿射变换的图像作为图像的输入,我该如何在deepstream的流水线中串起来,是不是需要自定义预处理的操作?因为nvinfer插件的预处理是内置的,不想用的话可能需要preprocess插件?
期待您的答疑
-
### Testing with the C270 HD Webcam Checked out from the Estore
-
Starting any sample currently logs messages like:
```
ModuleNotFoundError: No module named 'media_files_src_bin'
ModuleNotFoundError: No module named 'zeromq_sink'
ModuleNotFoundError: No module…
-
Do you have a plan to implement instance segmentation of yolov5, thanks.