BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171531Z
LOCATION:Hall B7 (1)\, B Block\, Level 7
DTSTART;TZID=Asia/Tokyo:20241205T153100
DTEND;TZID=Asia/Tokyo:20241205T154300
UID:siggraphasia_SIGGRAPH Asia 2024_sess135_papers_733@linklings.com
SUMMARY:From Sim-to-Real: Toward General Event-based Low-light Frame Inter
 polation with Per-scene Optimization
DESCRIPTION:Ziran Zhang (Zhejiang University, Shanghai Artificial Intellig
 ence Laboratory); Yongrui Ma (Chinese University of Hong Kong, Shanghai Ar
 tificial Intelligence Laboratory); Yueting Chen (Zhejiang University); Fen
 g Zhang (Shanghai Artificial Intelligence Laboratory); Jinwei Gu and Tianf
 an Xue (Chinese University of Hong Kong); and Shi Guo (Shanghai Artificial
  Intelligence Laboratory)\n\nVideo Frame Interpolation (VFI) is important 
 for video enhancement, frame rate up-conversion, and slow-motion generatio
 n. The introduction of event cameras, which capture per-pixel brightness c
 hanges asynchronously, has significantly enhanced VFI capabilities, partic
 ularly for high-speed, nonlinear motions. However, these event-based metho
 ds encounter challenges in low-light conditions, notably trailing artifact
 s and signal latency, which hinder their direct applicability and generali
 zation. Addressing these issues, we propose a novel per-scene optimization
  strategy tailored for low-light conditions. This approach utilizes the in
 ternal statistics of a sequence to handle degraded event data under low-li
 ght conditions, improving the generalizability to different lighting and c
 amera settings. To evaluate its robustness in low-light condition, we furt
 her introduce EVFI-LL, a unique RGB+Event dataset captured under low-light
  conditions. Our results demonstrate state-of-the-art performance in low-l
 ight environments. Project page: https://openimaginglab.github.io/Sim2Real
 /.\n\nRegistration Category: Full Access, Full Access Supporter\n\nLanguag
 e Format: English Language\n\nSession Chair: Changjian Li (University of E
 dinburgh)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_733&sess=sess135
END:VEVENT
END:VCALENDAR