koide3

@JSRGFJZ Since the paper is still under review, we have not made it public yet. If you are interested in the paper, please send a mail to [email protected] from your...

Hi @agn-7 , Can you run the nodelet in the standalone mode by changing "load" in the following line to "standalone"? https://github.com/koide3/hdl_people_tracking/blob/8c0d863e77207e7012217fcd16ce89f29a78ba67/launch/hdl_people_tracking.launch#L10

This package doesn't require LRFs. You can run it with a sole monocular camera. Can you check if some errors appear on the terminal?

Hi @ikunjin , Does "source /your/catkin_ws/devel/setup.bash && roscd hdl_people_tracking" work?

We tested several classifiers (SVM, AdaBoost, RandomForest) with Kidono's features. Since AdaBoost showed a bit better accuracy than the other ones, in this work, we are using AdaBoost instead of...

Since I did it a few years ago, and I left my former university last year, I'm not sure if I can find the SVM model... If you want, I...

We are using the features proposed in the following paper: Kidono et al., Pedestrian Recognition Using High-definition LIDAR, IV, 2011 You can extract the features from a point cloud with...

Here, you can find the dataset for pedestrian detection. It consists of KITTI, sydney, and our own dataset. To avoid some copyright problems, I put only the extracted features saved...

This package uses the standard cv::ml model, and you can replace it with any custom model you want to use. https://github.com/koide3/hdl_people_tracking/blob/master/src/kidono_human_classifier.cpp

Hi @agn-7 , As you pointed out, maybe this is due to the difference of the human shape in real and simulated data. You may track such still persons by...