DID-M3D

DID-M3D copied to clipboard

Hi I'm curious why the result `Car@BEV IoU=0.7[33.91,24.00,19.52] / Car@3D IoU=0.7[33.91 ,24.00, 19.52]` in this repo is better than the result `Car@BEV IoU=0.7[31.10,22.76, 19.50] / Car@3D IoU=0.7[25.38,17.07,14.06]` of original paper?...

zyrant

about data augmentation

1

Did this code change the depth when cropping?

zhenjiaa

question about the similar AP in KITTI test set and verification set

1

Hi, thanks for your great work and the great code.I'm having problems about your results. Use the pre-trained model you provide to test the AP results(Car@3D IoU=0.7: 25.38 17.07 14.06)...

yjy4231

about waymo dataset

2

Hi, thanks for your work and the code. The result on the waymo dataset is presented in the supplementary material in the paper and the result is also very good....

zyrant

depth dense map for test set

1

could you provide depth map for testing， not for training

sunhucheng

d = cv.imread('{}/{:0>6}.png'.format(self.dense_depth_dir, index), -1) / 256.

1

为什么读取的深度要除以256呢，因为后面的代码中： roi_depth_ind = (roi_depth > depth[i] - 3) & \ (roi_depth < depth[i] + 3) & \ (roi_depth > 0) depth[i]是一个绝对深度 roi_depth是由深度图d变换而来，他应该代表一个绝对深度，但是d已经除以256了 他似乎并不能代表绝对深度

aqr1961

Thank you for your work, which has a lot of reference for me. While using random crop and random scale augmentations ,the instance depth and the vis depth of the...

605789414

The Ap of batchsize 16 in one gpu

1

excuse me , i have retrained the model in one gpu with 16 batchsize, but i got the lower ap than paper. i have read the issue similar with my...

ray-peng

The difference between depth_dense and the velodyne file in the kitti data set

2

hello! have great insteresting about your work。For depth monitoring, you use a depth set trained on other networks(depth_dense) to monitor vis_depth, but I noticed that there is a velodyne file...

PopeyePxx

hello! i want know what is the depth_dense?how can i get it?

3

hello! have great insteresting about your work。 i want know what is the depth_dense?how can i get it?

Fanhuasijindd