Giters

SamsungLabs / imvoxelnet

[WACV2022] ImVoxelNet: Image to Voxels Projection for Monocular and Multi-View General-Purpose 3D Object Detection

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

can i uesd nuscenes_infos_val.coco.json file test iamge?

Bincle-Yin opened this issue · comments

commented

hi，i want used to nuscenes_infos_val.coco.json file test iamge，but not nuscenes_infos_val.pkl
e.g. config->base->dataset->newxxx.py

    test=dict(
        type=dataset_type,
        data_root=data_root,
        ann_file=data_root + 'nuscenes_infos_val_mono3d.coco.json',
        img_prefix=data_root,
        classes=class_names,
        pipeline=test_pipeline,
        modality=input_modality,
        test_mode=True,
        box_type_3d='Camera'))
evaluation = dict(interval=2)

Similar mmdetection3d->config->base->dataset->nus-mono3d.py

the idea is ok？

commented

Hi @Bincle-Yin ,

I'm not sure that the idea is ok, as we need 6 images per scene with intrinsic and extrinsic camera matrices. However this json file contains independent images without camera matrices.