artofstate
artofstate
@OuyangJunyuan hi,could you please share it with me ? My email is [email protected] Thank you very much!!!
I use RTX3090 with batch_size = 8 ,pytorch = 1.8
Sorry, Could you please show me in detail? How to get the position of every point with nv::Tensor type? In Cuda_BEVFusion/src/main.cpp, auto lidar_points = nv::Tensor::load(nv::format("%s/points.tensor", data), false); Then I'v tried...
@lqzhao 您好，看到您说points_in_boxes_cpu要求输入的 xyz 是bbox中心坐标？但是我看到源码中说 需要的是底部坐标？那是不是说明应该先gt_bbox_3d[:, 2] -= dims[:, 2] / 2，然后后边恢复gt_bbox_3d[:, 2] += dims[:, 2] / 2