דלג לתוכן (מקש קיצור 's')
אירועים

אירועים והרצאות בפקולטה למדעי המחשב ע"ש הנרי ומרילין טאוב

למידת ענני נקודות באמצעות מסלולים אקראיים עבור ניתוח מודלים תלת מימדיים
event speaker icon
עדי מסיכה (הרצאה סמינריונית למגיסטר)
event date icon
יום שלישי, 08.02.2022, 11:30
event location icon
Zoom Lecture: https://technion.zoom.us/my/chaimbaskin
event speaker icon
מנחה: Prof. Ayellet Tal
Point clouds are gaining prominence as a method for representing 3D shapes, but their irregular structure poses a challenge for deep learning methods. In this paper we propose CloudWalker, a novel method for learning 3D shapes using random walks. Previous works attempt to adapt Convolutional Neural Networks (CNNS) or impose a grid or mesh structure to 3D point clouds. This work presents a different approach for representing and learning the shape from a given point set. The key idea is to impose structure on the point set by multiple random walks through the cloud for exploring different regions of the 3D object. Then we learn a per-point and per-walk representation and aggregate multiple walk predictions at inference. Our approach achieves state-of-the-art results for two 3D shape analysis tasks: classification and retrieval. Furthermore, we propose a shape ambiguity indicator function that uses cross-walk and inter-walk variance measures to subdivid the shape space.