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:20231215T135500 DTEND;TZID=Australia/Melbourne:20231215T141000 UID:siggraphasia_SIGGRAPH Asia 2023_sess157_papers_441@linklings.com SUMMARY:Reconstruction of Machine-Made Shapes from Bitmap Sketches DESCRIPTION:Technical Papers, TOG\n\nIvan Puhachov (Universite de Montreal ; Huawei Technologies, Canada); Cedric Martens (Universite de Montreal); P aul G. Kry (McGill University; Huawei Technologies, Canada); and Mikhail B essmeltsev (Universite de Montreal)\n\nWe propose a method of reconstructi ng 3D machine-made shapes from bitmap sketches by separating an input imag e into individual patches and jointly optimizing their geometry. \nWe rely on two main observations:\n(1) human observers interpret sketches of man- made shapes as a collection of simple geometric primitives, and \n(2) sket ch strokes often indicate occlusion contours or sharp ridges between those primitives.\nUsing these main observations we design a system that takes a single bitmap image of a shape, estimates image depth and segmentation i nto primitives with neural networks, then fits primitives into the predict ed depth while determining occlusion contours and aligning intersections w ith input drawing via optimization.\nUnlike previous work, our approach do es not require additional input, annotation, or templates, and does not re quire retraining for a new category of man-made shapes. \nOur method produ ces triangular meshes that display sharp geometric features and are suitab le for downstream applications, such as editing, rendering, and shading.\n \nRegistration Category: Full Access\n\nSession Chair: Chi Wing Fu (The Ch inese University of Hong Kong) URL:https://asia.siggraph.org/2023/full-program?id=papers_441&sess=sess157 END:VEVENT END:VCALENDAR