Differenct coordinate setting in gt_bboxes and bboxes_preds

[nacayu]

Hello, thanks for your excellent work.
I have some questions about your work, In detr3d_head.py, the prediction coordinate is :

            tmp[..., 0:2] += reference[..., 0:2]
            tmp[..., 0:2] = tmp[..., 0:2].sigmoid()
            tmp[..., 4:5] += reference[..., 2:3]
            tmp[..., 4:5] = tmp[..., 4:5].sigmoid()
            tmp[..., 0:1] = (tmp[..., 0:1] * (self.pc_range[3] - self.pc_range[0]) + self.pc_range[0])
            tmp[..., 1:2] = (tmp[..., 1:2] * (self.pc_range[4] - self.pc_range[1]) + self.pc_range[1])
            tmp[..., 4:5] = (tmp[..., 4:5] * (self.pc_range[5] - self.pc_range[2]) + self.pc_range[2])

According to the above setting, the tmp[...,:2] is x,y, and the 4th dim is z.
But in ground truth setting:
The gt_bboxes_3d[...,:3] is x,y,z which is conflict with the prediction bboxes of dimension setting.

        if self.with_velocity:
            gt_velocity = info['gt_velocity'][mask]
            nan_mask = np.isnan(gt_velocity[:, 0])
            gt_velocity[nan_mask] = [0.0, 0.0]
            gt_bboxes_3d = np.concatenate([gt_bboxes_3d, gt_velocity], axis=-1)

How to explain it?

[a1600012888]

Hi, see the normalize and denonormalize function here: https://github.com/WangYueFt/detr3d/blob/main/projects/mmdet3d_plugin/core/bbox/util.py#L26

There are processes where we convert the format of the box using there two functions.

[nacayu]

@a1600012888 Thanks

[chyohoo]

hello, in normalize_bbox, why log the w,h,l for loss calculating?