turboxin

Hi all, thanks for releasing this wonderful project！ I 'm wondering what's your recent development plan for new models or methods?

> The provided checkpoint is finetuned from the detection task. I will update more experiments. Hi, I'm wondering how to finetune the model on detection task and why would that...

@muzi2045 Hi Muzi，for pretraing your 16 beam model with kitti and nuscene, did you use the original 64/32 beam data , or downsample them to 16 beam? Thank you in...

@muzi2045 thank you very much!

@muzi2045 hello！You mentioned that inference_time is between 30ms~50ms, may I ask what GPU are you using? Could you please also provide some quantitative performance data on your results on 16...

Hi @Yang7879 ，I‘m wondering is the gredient of Hungarian assignment really necessary? Can we train the network without this gradient?

@tomztyang Looking forward to nuscenes config~ Thanks a lot!

> > @lilanxiao ,hi.thanks for your great work,it's help me a lot. In the repository, Is the coordinates' orientation like@a-akram-98 said? `x` to `left` ` y` to `forward` `z` to...

@wangjksjtu hi, would it be possible to provide the intrinsic and extrinsic paramenters of the camera and transformation matrix between camera and lidar, it would be really helpful and much...

Really sorry to bother you @poodarchu, is there any update about PointRCNN release time? Thanks a lot.