using image segmentation result instead of object detection

ZrrSkywalker / MonoDETR

[ICCV 2023] The first DETR model for monocular 3D object detection with depth-guided transformer

344 stars 29 forks source link

using image segmentation result instead of object detection #47

Open lturing opened 10 months ago

lturing commented 10 months ago

according to "The pixels within the same 2D bounding box are assigned with the same depth label of the corresponding object", so if using image segmentation, the result will better?