DOCTR: Disentangled Object-Centric Transformer for Point Scene Understanding.

Xiaoxuan Yu, Hao Wang, Weiming Li, Qiang Wang, SoonYong Cho, Younghun Sung

Browse the full AAAI paper archive.