detic_ros

ROS package for Detic. Run on both CPU and GPU, GPU is way performant, but work fine also with CPU (take few seconds to process single image).

example of custom vocabulary. Left: default (lvis), Right: custom ('bottle,shoe')

example of three dimensional pose recognition for cups, bottles, and bottle caps.

step1: build docker container

Ofcourse you can build this pacakge on your workspace and launch as normal ros package. But for those using CUDA, the following docker based approach might be safer and easy.

Prerequsite: You need to preinstall nvidia-container-toolkit beforehand. see (https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html)

Build docker image

git clone https://github.com/HiroIshida/detic_ros.git
cd detic_ros
docker build -t detic_ros .

step2: launch Detic-segmentor node

Example for running node on pr1040 (please replace pr1040 by you robot hostname or localhost):

python3 run_container.py -host pr1040 -mount ./launch -name sample.launch \
    out_debug_img:=true \
    out_debug_segimg:=false \
    compressed:=false \
    device:=auto \
    input_image:=/kinect_head/rgb/image_color

The minimum necessary argument of run_container.py is host, mount and name:

host: host name or IP address
mount: launch file or launch file's directory path that will be mounted inside the container. In this example, launch file directory of this repository is mounted.
name: launch file name that will be searched from mounted file or directory Also, you can specify launch args as the roslaunch command (e.g. out_debug_img:=true). This launch args must come after the above three args.

Another example for running three dimensional object pose detection using point cloud filtered by segmentation.

python3 run_container.py -host pr1040 -mount ./launch -name sample_detection.launch \
    debug:=true \
    vocabulary:=custom \
    custom_vocabulary:=bottle,cup

Or rosrun detic_ros run_container.py if you catkin build this package on the hosting computer side. As in this example, by putting required sub-launch files inside the directory that will be mounted on, you can combine many node inside the container.

Note:

On custom vocabulary: if you want to limit the detected instances by custom vocabulary, please set launch args to vocabulary:='custom' custom_vocabulary:='bottle,shoe'.
On model types: Detic is trained in different model types. In this repository you can try out all of the real-time models using the model_type parameter.
On real-time performance: For higher recognition frequencies turn off all debug info, run on GPU, decompress topics locally, use smaller models (e.g. res50), and avoid having too many classes in the frame (by e.g. setting a custom vocabulary or higher confidence thresholds). The sample_detection.launch with default parameters handles all of this, yielding object bounding boxes at around 10Hz.

step3a (Subscribe from node in step3 and do anything you want)

Example for using the published topic from the node above is masked_image_publisher.py. This will be helpful for understanding how to apply SegmentationInfo message to a image. The test file for this example also might be helpful.

step3b (Service call)

See definition of srv/DeticSeg.srv

ROS node information

~input_image (sensor_msgs/Image)
- Input image
~debug_image (sensor_msgs/Image)
- debug image
~debug_segmentation_image (sensor_msgs/Image with 32SC1 encoding)
- Say detected class number is 14, ~segmentation_image in grayscale image is almost completely dark and not good for debugging. Therefore this topic scale the value to [0 ~ 255] so that grayscale image is human-friendly.
~segmentation_info (detic_ros/SegmentationInfo)
- Published when use_jsk_msgs is false. Includes the class name list, confidence score list and segmentation image with 32SC1 encoding. The image is filled by 0 and positive integers indicating segmented object number. These indexes correspond to one plus those of class name list and confidence score list. For example, an image value of 2 corresponds to the second (index=1) item in the class name and score list. Note that the image value of 0 is always reserved for the 'background' instance.
~segmentation (sensor_msgs/Image)
- Published when use_jsk_msgs is true. Includes the segmentation image with 32SC1 encoding.
~detected_classes (jsk_recognition_msgs/LabelArray)
- Published when use_jsk_msgs is true. Includes the names and ids of the detected objects. In the same order as ~score.
~score (jsk_recognition_msgs/VectorArray)
- Published when use_jsk_msgs is true. Includes the confidence score of the detected objects. In the same order as ~detected_classes.

As for rosparam, see node_cofig.py.

Running without roscore to batch processing a bag file

rosrun detic_ros batch_processor.py path/to/bagfile

See source code for the options.

HiroIshida / detic_ros

readme