Closed huyusheng123 closed 3 years ago
Hi @yzrobot ,
After considering, I have some idea for libsvm, now I want to use LIBSVM to train my dataset, but I fail to obtain the file named "svm-train.c", Could you please provide this file to me?
best wishes thanks.
Hi, I have not had the chance to test it on the Velodyne HDL-64. But IMHO, if you use the offline version object3d_detector.launch
then you probably need to retrain your SVM classifier with the Velodyne HDL-64 data, but if you use the online version object3d_detector_ol.launch
then you can get the SVM classifier more easily, you just need to uncomment line 815 and 816 https://github.com/yzrobot/online_learning/blob/Kinetic/object3d_detector/src/object3d_detector_ol.cpp
.
The source of LIBSVM is here https://github.com/cjlin1/libsvm
Cheers, Zhi
Hi @yzrobot ,
I am happy to hear from you, now It means that I can get the SVM classifier for my bag file which have the same topics after changing the related parameter in object3d_detector_ol.cpp?
the second question: now I can get the eleven parameters of label in per PCD file by using your package (https://github.com/yzrobot/cloud_annotation_tool), If I want to re-train my SVM classifier with the Velodyne HDL-64 data, What should I do next?
best wishes thanks.
@huyusheng123 Hi: I meet the same problem by using 64-line LIDAR. So I want to know if you have a solution to fix this problem. Best wishes.
Hi, you could try this: https://github.com/yzrobot/cloud_annotation_tool/tree/devel. Please modify the source code as needed. Cheers.
@huyusheng123 Hi, I am sorry to bother you.
I meet the same problem and I want to know if you have a solution to train the SVM classifier?
Any help is much appreciated:)
Hi,
I think that it does not work with the equipmen of Velodyne HDL-64 because the different feature extract from points, so could you please tell me how to use LIBSVM to obtain six features of human from points.
best wishes Thanks.