We propose 3DETR, an end-to-end Transformer based object detection model for 3D point clouds . Compared to existing detection methods that employ a number of 3D-specific inductive biases, 3DEER requires minimal modifications to the Transformer block . Through extensive experiments, we show 3DDER outperforms the well-established and highly optimized VoteNet baselines on the challenging ScanNetV2 dataset by 9.5% . We show that 3DEDR is applicable to 3D tasks beyond detection, and can serve as a building block for future research .

Author(s) : Ishan Misra, Rohit Girdhar, Armand Joulin

Links : PDF - Abstract

Code :

Keywords : d - detection - transformer - block - object -

Leave a Reply

Your email address will not be published. Required fields are marked *