maudzung / SFA3D

Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds (The PyTorch implementation)
https://github.com/maudzung/Super-Fast-Accurate-3D-Object-Detection
MIT License
1k stars 271 forks source link

Cannot get detections on VLP 16 lidar data #70

Closed Arshad9544 closed 2 years ago

Arshad9544 commented 2 years ago

I tried using the pre-trained model on data from VLP 16 Lidar but couldn't get any detections. Will it work if I change the kitti config file or how the BEV map is formed. Please help. Thank you

maudzung commented 2 years ago

Point clouds collected from 16-channel LiDARs are quite sparse, so you need to retrain the model to get better accuracy.