You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
打印信息
================ Start to Convert ================
================ Start to Copy Raw Data ================
================ Start to Generate Label ================
================ Start to Generate Calibration Files ================
15627 15627
================ Start to Generate ImageSet Files ================
您好,在DAIR-V2X-V数据集,通过下面的命令转为kitii格式:
python dair2kitti.py --source-root ./data/DAIR-V2X-V/single-vehicle-side --target-root ./data/DAIR-V2X-V/single-vehicle-side --split-path ./data/split_datas/single-vehicle-split-data.json --label-type camera --sensor-view vehicle
打印信息
================ Start to Convert ================
================ Start to Copy Raw Data ================
================ Start to Generate Label ================
================ Start to Generate Calibration Files ================
15627 15627
================ Start to Generate ImageSet Files ================
查看calib中的相机标定文件,比如 000000.txt
P2: 3996.487567 0.0 955.58618 0.0 0.0 3963.430994 527.646219 0.0 0.0 0.0 1.0 0.0
P2: 3996.487567 0.0 955.58618 0.0 0.0 3963.430994 527.646219 0.0 0.0 0.0 1.0 0.0
P2: 3996.487567 0.0 955.58618 0.0 0.0 3963.430994 527.646219 0.0 0.0 0.0 1.0 0.0
P2: 3996.487567 0.0 955.58618 0.0 0.0 3963.430994 527.646219 0.0 0.0 0.0 1.0 0.0
R0_rect: 1 0 0 0 1 0 0 0 1
Tr_velo_to_cam: 0.006283 -0.999979 -0.001899 -0.298036 -0.005334 0.001865 -0.999984 -0.666812 0.999966 0.006293 -0.005322 -0.516927
Tr_velo_to_cam: 0.006283 -0.999979 -0.001899 -0.298036 -0.005334 0.001865 -0.999984 -0.666812 0.999966 0.006293 -0.005322 -0.516927
查看lable_2中的标签,比如 000000.txt
Car 0 0 0.33888581543844903 0 527.938232 69.723068 637.4556269999999 0.850836 4.337498 2.073565 -9.601712831407 0.8624079931420001 32.383280568744 1.615145
看起来转换应该梅问题的,但是,在可视化3D检测框时发现部分标签不准确,而且不准确的标签占比较大。
比如:000187.jpg
比如:000141.jpg
比如:000184.jpg
比如:000112.jpg
比如:000010.jpg
请问标签确实会存在偏差是吗
The text was updated successfully, but these errors were encountered: