Closed bpinaya closed 6 years ago
Hmm thanks Benjamin, let me see if I can locate this, otherwise we may need to re-create the nodelet from the gstCamera sources included with jetson-inference. These nodelets were an experiment to get better performance anyhow. Turns out we needed to author our own I/O nodelets as well.
You may also be interested, we now have caffe_ros node available under the github.com/NVIDIA-Jetson account. It loads caffe models with TensorRT and the node portion is more current than what I have here.
From: Benjamin Pinaya [mailto:notifications@github.com] Sent: Friday, September 22, 2017 5:24 AM To: dusty-nv/ros_deep_learning ros_deep_learning@noreply.github.com Cc: Dustin Franklin dustinf@nvidia.com; Mention mention@noreply.github.com Subject: [dusty-nv/ros_deep_learning] Failed to load nodelet /gst_camera of type [ros_jetson_video/gst_camera] (#3)
Hi there @dusty-nvhttps://github.com/dusty-nv , in the launch file you include a node which I assume is the node that takes care of the gst_camera. Do you plan to release such node? Otherwise there is this error:
[ERROR] [1506071741.153780987]: Failed to load nodelet [/gst_camera] of type [ros_jetson_video/gst_camera] even after refreshing the cache: According to the loaded plugin descriptions the class ros_jetson_video/gst_camera with base class type nodelet::Nodelet does not exist. Declared types are ros_deep_learning/ros_imagenet
[ERROR] [1506071741.153964994]: The error before refreshing the cache was: According to the loaded plugin descriptions the class ros_jetson_video/gst_camera with base class type nodelet::Nodelet does not exist. Declared types are ros_deep_learning/ros_imagenet
[FATAL] [1506071741.154335455]: Failed to load nodelet '/gst_cameraof type
ros_jetson_video/gst_camerato manager
standalone_nodelet'
[gst_camera-4] process has died [pid 32494, exit code 255, cmd /opt/ros/kinetic/lib/nodelet/nodelet load ros_jetson_video/gst_camera standalone_nodelet ~image_raw:=/image_raw __name:=gst_camera __log:=/home/ubuntu/.ros/log/92f488dc-9f76-11e7-b5c5-00044b633dff/gst_camera-4.log].
log file: /home/ubuntu/.ros/log/92f488dc-9f76-11e7-b5c5-00044b633dff/gst_camera-4*.log
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/dusty-nv/ros_deep_learning/issues/3, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AOpDK80uUsmsSplAYOwOInnB4EYHb-ELks5sk3ysgaJpZM4Pgcln.
Oh thanks! I'll check that one out. Closing this one.
Hey @dusty-nv , did you find the nodelet by any chance? You used the files in the gst-camera from the jetson-inference to build it right? I think I'll try to recreate the nodelet, I tried the jetson caffe_ros repo but it's not working and haven't gotten their launch files with the gscam node to work.
Did either of you (re)create this nodelet?
Hey there, I ended up doing an implementation myself but can disclose it since it was done for a company, but it basically resumes to get the image from a rostopic, converting it to your format and pass it to the serialized TensorRT Network. But I think it's better to use the direct gstcam or v4l to get the images.
Hi there @dusty-nv , in the launch file you include a node which I assume is the node that takes care of the gst_camera. Do you plan to release such node? Otherwise there is this error: