praveen-palanisamy / multiple-object-tracking-lidar

C++ implementation to Detect, track and classify multiple objects using LIDAR scans or point cloud
MIT License
796 stars 229 forks source link

What are the classes in which the code classifies the objects? Any accuracy metric available? What is the speed you are getting? #43

Closed Surajj4837-th closed 3 years ago

Surajj4837-th commented 3 years ago

What are the classes in which the code classifies the objects? Any accuracy metric available? What is the speed you are getting?

praveen-palanisamy commented 3 years ago

Hi @Surajj4837-th. Missed this question from you until now. Please see my answers below:

What are the classes in which the code classifies the objects? Any

Any static/dynamic object is detected and tracked. The clustering of pointclouds into an object happens in an unsupervised way so there's no "class labels" associated with them.

Any accuracy metric available?

No. It's been mainly used on live pointcloud streams and wasn't benchmarked on a static dataset which has annotations to get such metrics.

What is the speed you are getting?

IIRC, the results were published >=30FPS. It depends on the number of points in the pointcloud and the hardware this is running

Surajj4837-th commented 3 years ago

Thanks for the update.

boazMgm commented 2 years ago

Hi, Is there a way to track only dynamic objects? Thanks a lot.