SPengLiang/DID-M3D
[ECCV 2022] DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection.
PythonMIT
Issues
- 0
d = cv.imread('{}/{:0>6}.png'.format(self.dense_depth_dir, index), -1) / 256. TypeError: unsupported operand type(s) for /: 'NoneType' and 'float'
#36 opened by yinianwujide - 1
depth dense map for test set
#35 opened by sunhucheng - 8
how to get the depth_dense
#3 opened by MillX2021 - 3
- 1
- 0
- 1
Can you provide the visualization code
#30 opened by yjy4231 - 1
The Ap of batchsize 16 in one gpu
#29 opened by ray-peng - 2
The difference between depth_dense and the velodyne file in the kitti data set
#28 opened by yangfan293 - 5
Non-deterministic algorithm
#18 opened by alsrb4446 - 2
For the submission on KITTI 3D OD benchmark
#27 opened by mingyuShin - 3
- 1
about attribute depth
#25 opened by yjy4231 - 2
Visual depth affine transform
#24 opened by mingyuShin - 4
test set AP
#22 opened by yjy4231 - 4
Visualization
#23 opened by Xiaolong-RRL - 1
about data augmentation
#13 opened by zhenjiaa - 2
vis_depth = (-vis_depth).exp() ?
#20 opened by Xiaolong-RRL - 2
Is ” visible depth” = 2D projected Lidar depth?
#21 opened by Senwang98 - 2
The Monocular setting and performance comparison
#19 opened by JCZ404 - 2
There is error between eval mAP during training and inference mAP after training
#17 opened by Senwang98 - 2
Run inference on a video frame by frame
#16 opened by y0088766 - 2
about waymo dataset
#11 opened by zyrant - 2
the difference between of original paper and this repo
#14 opened by zyrant - 3
- 1
Reproducibility
#15 opened by alsrb4446 - 6
the AP of train model
#5 opened by MillX2021 - 1
trained on single gpu
#7 opened by KuofengGao - 7
- 7
Question about decoupled instance depth
#9 opened by Senwang98 - 3
Question about scale_box2d_masked.
#8 opened by kaixinbear - 3
- 2
About the license for this model
#4 opened by PINTO0309 - 2
- 4
Question about submission
#1 opened by konyul