Karl
Karl
and when my video_multiframe>1 ,it can process my video ,the output is same as input
hey,have you solved the problem?How can I evaluate the autoencoder on my own data?thanks a lot!
> there is no file that works out of the box. But you can check this file to get an idea for how to use open3d to visualize detections https://github.com/tianweiy/CenterPoint/blob/master/tools/visual.py...
sorry to disturb you tianwei, I am trying single_infernece_ros.py in your code, and when processing the point cloud data,I meet this error: [ERROR] [1648210357.932211]: bad callback: Traceback (most recent call...
I solved this by changing the max_voxel_num in the config file from a list to a integer. The model trained on nuscenes dataset doesn't work well for my data collected...
Thanks, I will try it, did you provide pretrained models on Waymo?
btw,my test data is a continuous ploud cloud data stream record by 16 line Velodyne lidar in 10 mins,is this satisfy the "motion cues"?
I will try Waymo first I noticed [multi_sweep_inference_ros.py](https://github.com/tianweiy/CenterPoint/blob/master/tools/multi_sweep_inference_ros.py) subscribe odom topic, what does it refer to?
and my numpy version is 1.23.5
Thanks, the glog version is 0.4.0, and gflags 2.2.2 I met this error on NVIDIA JETSON ORIN, while it did not appear on a x86 computer. I have checked that...