跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.22207 · v1

ARG-Avatar: Animation-Rendering Guidance for Efficient Feedforward Avatar Reconstruction

首次提交 2026-11-17 12:45
版本更新 2026-11-17 12:45
雷达观测 2026-11-18 19:43
arXiv ↗PDF ↗

摘要

Reconstructing animatable 3D human avatars from images is important for applications such as gaming and augmented reality. Feedforward methods make avatar reconstruction accessible by reducing inference time to a few seconds, but recent approaches rely heavily on scaling model capacity, training data, and computation. We ask whether such scaling is necessary. To this end, we introduce ARG-Avatar, an efficient feedforward framework with only 68M trainable parameters that achieves SOTA reconstruction quality. Our key insight is that 3D animation and rendering already provide strong geometric correspondences that existing methods largely attempt to learn from data. In light of this, we design the Foreground Avatar Coordinate RoPE attention (FACRoPE) and Intermediate Token Rendering (ITR) pipeline to guide avatar tokens to attend to their corresponding patches. Experiments across multiple datasets show that ARG-Avatar consistently outperforms prior methods when using two or more input images. Compared with LHM++, ARG-Avatar uses approximately 9×\times fewer trainable parameters, 4.8×\times fewer training sequences, and 14×\times less training compute, while running 6×\times faster and achieving a 2.7 dB PSNR improvement on the out-of-distribution NeuMan benchmark.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-17 12:45 · 最新观测版本

ARG-Avatar: Animation-Rendering Guidance for Efficient Feedforward Avatar Reconstruction

Reconstructing animatable 3D human avatars from images is important for applications such as gaming and augmented reality. Feedforward methods make avatar reconstruction accessible by reducing inference time to a few seconds, but recent approaches rely heavily on scaling model capacity, training data, and computation. We ask whether such scaling is necessary. To this end, we introduce ARG-Avatar, an efficient feedforward framework with only 68M trainable parameters that achieves SOTA reconstruction quality. Our key insight is that 3D animation and rendering already provide strong geometric correspondences that existing methods largely attempt to learn from data. In light of this, we design the Foreground Avatar Coordinate RoPE attention (FACRoPE) and Intermediate Token Rendering (ITR) pipeline to guide avatar tokens to attend to their corresponding patches. Experiments across multiple datasets show that ARG-Avatar consistently outperforms prior methods when using two or more input images. Compared with LHM++, ARG-Avatar uses approximately 9×\times fewer trainable parameters, 4.8×\times fewer training sequences, and 14×\times less training compute, while running 6×\times faster and achieving a 2.7 dB PSNR improvement on the out-of-distribution NeuMan benchmark.

查看此版本解读 →查看此版本原文 ↗