»ã±¨±êÌâ (Title)£ºCP3: Channel Pruning Plug-in for Point-based Networks£¨»ùÓÚµãµÄÍøÂçͨ·½¨¼ô£©

»ã±¨ÈË (Speaker)£º Éò³¬Ãô ¸±½ÌÊÚ£¨»ª¶«Ê¦·¶´óѧ£©

»ã±¨¹¦·ò (Time)£º2023Äê11ÔÂ2ÈÕ(ÖÜËÄ) 8:30

»ã±¨µØÖ· (Place)£ºÐ£±¾²¿ F420

Ô¼ÇëÈË(Inviter)£ºÎÂÖÇæ¼

Ö÷°ì²¿ÃÅ£ºÀíѧԺÊýѧϵ

»ã±¨ÌáÒª£º In this talk, we propose a model called CP3, which is a Channel Pruning Plug-in for Point-based network. CP3 is elaborately designed to leverage the characteristics of point clouds and PNNs in order to enable 2D channel pruning methods for PNNs. Specifically, it presents a coordinate-enhanced channel importance metric to reflect the correlation between dimensional information and individual channel features, and it recycles the discarded points in PNN¡¯s sampling process and reconsiders their potentially-exclusive information to enhance the robustness of channel pruning. Experiments on various PNN architectures show that CP3 constantly improves state-of-the-art 2D CNN pruning approaches on different point cloud tasks.