BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260417T190110Z
LOCATION:West Building\, Rooms 109-110
DTSTART;TZID=America/Los_Angeles:20250814T140000
DTEND;TZID=America/Los_Angeles:20250814T153000
UID:siggraph_SIGGRAPH 2025_sess173@linklings.com
SUMMARY:ML in Production
DESCRIPTION:Towards automated corrections in video-driven animation transf
 er\n\nWe improve Digital Domain's video-driven animation transfer techniqu
 e by introducing automatic corrections as a post-process optimization. We 
 minimize the difference between our face swap model output (extended for l
 ight invariance) and predicted animation parameters in a differentiable pi
 peline. Ou...\n\n\nJose Serra and Lucio Moser (Digital Domain)\n----------
 -----------\nMachine Learning Meets Lighting: Using Depth Estimation To Bu
 ild The Light Rigs\n\nThis paper describes the techniques we use to build 
 the complex light rigs in the Sony Pictures Imageworks lighting pipeline. 
 Our process is semi-automatic and removes manual labour from the artists.\
 n\n\nSergey Shlyaev (Sony Pictures Imageworks, Canada)\n------------------
 ---\nAniDepth : Anime In-between Diffusion using Depth-guided Warped Line-
 art\n\nWe propose AniDepth, a novel anime in-betweening method using a vid
 eo diffusion model enhanced by converting anime illustrations into depth m
 aps. Guided by line-arts, our approach interpolates depth maps and colors 
 to boost fidelity, temporal smoothness, and performance while seamlessly i
 ntegrating ...\n\n\nSosui Koga (Nara Institute of Science and Technology);
  Hiroyuki Kubo (Chiba University); Seitaro Shinagawa, Yuki Fujimura, and K
 azuya Kitano (Nara Institute of Science and Technology); Akinobu Maejima (
 OLM Digital, Inc.; IMAGICA GROUP Inc.); and Takuya Funatomi and Yasuhiro M
 ukaigawa (Nara Institute of Science and Technology)\n---------------------
 \nTowards AI-Driven 3D Creation at the Speed of Thought\n\nHow can generat
 ive AI seamlessly integrate into professional 3D workflows? This talk expl
 ores how vision-language models (VLMs) can automate tedious editing tasks,
  generate structured 3D scenes, and perform object placement while preserv
 ing editability. I’ll discuss key findings, open challeng...\n\n\nIan Huan
 g (Stanford University)\n\nInterest Area: Arts & Design, New Technologies,
  Production & Animation, Research & Education\n\nRecording: Livestreamed, 
 Not Livestreamed, Recorded, Not Recorded\n\nKeyword: Animation, Artificial
  Intelligence/Machine Learning, Capture/Scanning, Computer Vision, Generat
 ive AI, Geometry, Image Processing, Lighting, Pipeline Tools and Work, Ren
 dering\n\nRegistration Category: Full Conference, Virtual Access, Thursday
 \n\nSession Chair: Nora Wixom (Apple)
END:VEVENT
END:VCALENDAR