Post

Neural Parametric Gaussian Avatars πŸ”΄ πŸ”΄ πŸ”΄ ( NPGA )

The creation of high-fidelity digital human heads is crucial for integrating virtual elements into daily life, demanding photo-realism and real-time rendering.

Neural Parametric Gaussian Avatars (NPGA) use a data-driven approach with 3D Gaussian splatting for efficient rendering and topological flexibility, conditioning avatar dynamics on neural parametric head models (NPHM).

Evaluations on the NeRSemble dataset show NPGA outperforms previous methods in self-reenactment tasks and demonstrates accurate animations from monocular videos.

πŸ§™Paper Authors: Minghan Qin1, Wanhua Li2†, Jiawei Zhou1, Haoqian Wang1†, Hanspeter Pfister2 ( indicates equal contribution, † means Co-corresponding author) 1Tsinghua University, 2Harvard University

Translate to Korean

고좩싀도 디지털 휴먼 ν—€λ“œμ˜ 생성은 가상 μš”μ†Œλ₯Ό 일상 μƒν™œμ— ν†΅ν•©ν•˜λŠ” 데 맀우 μ€‘μš”ν•˜λ©°, 포토 λ¦¬μ–Όλ¦¬μ¦˜κ³Ό μ‹€μ‹œκ°„ λ Œλ”λ§μ΄ μš”κ΅¬λ©λ‹ˆλ‹€.

NPGA(Neural Parametric Gaussian Avatars)λŠ” 효율적인 λ Œλ”λ§κ³Ό ν† ν΄λ‘œμ§€ μœ μ—°μ„±μ„ μœ„ν•΄ 3D κ°€μš°μŠ€ μŠ€ν”Œλž˜νŒ…κ³Ό ν•¨κ»˜ 데이터 기반 μ ‘κ·Ό 방식을 μ‚¬μš©ν•˜μ—¬ NPHM(Neural Parametric Head Model)μ—μ„œ 아바타 역학을 μ‘°μ ˆν•©λ‹ˆλ‹€.

NeRSemble 데이터 μ„ΈνŠΈμ— λŒ€ν•œ 평가에 λ”°λ₯΄λ©΄ NPGAλŠ” 자체 μž¬μ—° μž‘μ—…μ—μ„œ 이전 방법을 λŠ₯κ°€ν•˜λ©° λ‹¨μ•ˆ λΉ„λ””μ˜€μ˜ μ •ν™•ν•œ μ• λ‹ˆλ©”μ΄μ…˜μ„ λ³΄μ—¬μ€λ‹ˆλ‹€.

This post is licensed under CC BY 4.0 by the author.