Test with ouster lidar

[irmuun20]

I would like to test model on ouster lidar data. How can I set the following parameters? what is the difference between two X_MIN, and X_CENTER_MIN etc and what is X_STEP?

// pillar size 
#define X_STEP 0.32f
#define Y_STEP 0.32f
#define X_MIN -74.88f
#define X_MAX 74.88f
#define Y_MIN -74.88f
#define Y_MAX 74.88f
#define Z_MIN -2.0f
#define Z_MAX 4.0f

#define X_CENTER_MIN -80.0f
#define X_CENTER_MAX 80.0f
#define Y_CENTER_MIN -80.0f
#define Y_CENTER_MAX 80.0f
#define Z_CENTER_MIN -10.0f
#define Z_CENTER_MAX 10.0f

#define PI 3.141592653f
// paramerters for preprocess
#define BEV_W 468
#define BEV_H 468
#define MAX_PILLARS 32000 //20000 //32000
#define MAX_PIONT_IN_PILLARS 20
#define FEATURE_NUM 10
#define PFE_OUTPUT_DIM 64
#define THREAD_NUM 4

[irmuun20]

[irmuun20]

With ouster sample data, it is detecting trees as people.

[irmuun20]

Used ouster lidar Signal 1 and Signal 2 Values instead of intensity and elongation (waymo):

intensity = np.tanh(Signal1 / 255.0)
elongation = np.tanh(Signal2 / 255.0)

and got the following results: cars are detected (most of them) but people are barely being detected: There is a person on the road that is very close to lidar but not being detected.

[HaohaoNJU]

X_MIN， XMAX are used to define point cloud range in preprocess and BEV size, where $H{BEV} = (X{MAX} - X{MIN})/X_{STEP}$

X_CENTER_MIN& X_CENTER_MAX define ranges that are used when making post processing where objs out of that range are not detected

[irmuun20]

What variables should be changed when one uses different lidar like ouster 2(128)? Thanks

[HaohaoNJU]

Hi, I think it's due to the data difference, the model I provide is trained from waymo dataset velodyne 64 lidar. Maybe you should train a new model with your new dataset, and deploy it as TRT model, then run inference, in README I provide the details on how to deploy your own model.

BW, Hao

irmuun20 @.***> 于2023年3月13日周一 08:20写道：

What variables should be changed when one uses different lidar like ouster 2(128)? Thanks

— Reply to this email directly, view it on GitHub https://github.com/HaohaoNJU/CenterPoint/issues/29#issuecomment-1465343633, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGF2XJM2JYCH64IHFUZMKBLW3ZR6VANCNFSM6AAAAAAVTEU4DA . You are receiving this because you commented.Message ID: @.***>

[irmuun20]

Thanks for the suggestion but at the moment, I don't have annotated cloud point data collected by my lidar. So, for now I will use waymo setting as default, and when I get my own data, I will train model as you suggested.

[mariamsamiiiir]

Hello , I need to know if signal2 is the refelctivitty or the range in ouster lidar because I want to map it to elongation like you have done but i don't it is which value. Thank you