Jiajun Deng
Jiajun Deng
The box is not annotated to match a certain phrase, but the whole sentence.
Hi @mc171819 , This error hasn't occurred to me. I suggest you run the code with the docker image I provide.
Hi, I haven't tried transfer learning in point clouds domains, since there are domain gaps for point clouds captured by different sensors and in different scenes. But I recommend you...
Hi @linhuixiao, Thanks for your interest in TransVG++. This paper is under review now. It may take several days before the code is made available. Best, Jiajun
Hi, You can run it twice, or try to use OpenPCDet for reproducing. I have added Voxel R-CNN into OpenPCDet now. Best Regards, Jiajun Deng
I have never trained it for more than 80 epochs. Have you used the same batch size setting as mine? That's an interesting finding. Thanks for your comments.
Hi， You can try it again. Have you ever tested other epochs?
Hi, I mean that maybe you should train it again. My results are as follows: Car [email protected], 0.70, 0.70: bbox AP:90.7833, 89.8625, 89.3937 bev AP:90.3749, 88.4831, 87.6475 3d AP:89.6334, 85.2028,...
I also use 8GPUs with batchsize=4 per GPU. The results are weird. I'll pull this code and train it again.
I have finished the README now. Maybe you can also train it again with the docker image I provide.