BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240214T070250Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T134500
DTEND;TZID=Australia/Melbourne:20231215T135500
UID:siggraphasia_SIGGRAPH Asia 2023_sess157_papers_819@linklings.com
SUMMARY:Learning Gradient Fields for Scalable and Generalizable Irregular 
 Packing
DESCRIPTION:Technical Papers, TOG\n\nTianyang Xue (Shandong University), M
 ingdong Wu (Peking University), Lin Lu and Haoxuan Wang (Shandong Universi
 ty), and Hao Dong and Baoquan Chen (Peking University)\n\nThe packing prob
 lem, also known as cutting or nesting, has diverse applications in logisti
 cs, manufacturing, layout design, and atlas generation. It involves arrang
 ing irregularly shaped pieces to minimize waste while avoiding overlap. Re
 cent advances in machine learning, particularly reinforcement learning, ha
 ve shown promise in addressing the packing problem. In this work, we delve
  deeper into a novel machine learning-based approach that formulates the p
 acking problem as conditional generative modeling. To tackle the challenge
 s of irregular packing, including object validity constraints and collisio
 n avoidance, our method employs the score-based diffusion model to learn a
  series of gradient fields. These gradient fields encode the correlations 
 between constraint satisfaction and the spatial relationships of polygons,
  learned from teacher examples. During the testing phase, packing solution
 s are generated using a coarse-to-fine refinement mechanism guided by the 
 learned gradient fields. To enhance packing feasibility and optimality, we
  introduce two key architectural designs: multi-scale feature extraction a
 nd coarse-to-fine relation extraction. We conduct experiments on two typic
 al industrial packing domains, considering translations only. Empirically,
  our approach demonstrates spatial utilization rates comparable to, or eve
 n surpassing, those achieved by the teacher algorithm responsible for trai
 ning data generation. Additionally, it exhibits some level of generalizati
 on to shape variations. We are hopeful that this method could pave the way
  for new possibilities in solving the packing problem.\n\nRegistration Cat
 egory: Full Access\n\nSession Chair: Chi Wing Fu (The Chinese University o
 f Hong Kong)
URL:https://asia.siggraph.org/2023/full-program?id=papers_819&sess=sess157
END:VEVENT
END:VCALENDAR