paper-with-me

홈 › Papers

AMG: Avatar Motion Guided Video Generation

2024-09-02 · Zhangsihao Yang, Mengyi Shan, Mohammad Farazi, Wenhui Zhu, Yanxi Chen, Xuanzhao Dong, Yalin Wang

Human video generation task has gained significant attention with the advancement of deep generative models. Generating realistic videos with human movements is challenging in nature, due to the intricacies of human body topology and sensitivity to visual artifacts. The extensively studied 2D media generation methods take advantage of massive human media datasets, but struggle with 3D-aware control; whereas 3D avatar-based approaches, while offering more freedom in control, lack photorealism and cannot be harmonized seamlessly with background scene. We propose AMG, a method that combines the 2D photorealism and 3D controllability by conditioning video diffusion models on controlled rendering of 3D avatars. We additionally introduce a novel data processing pipeline that reconstructs and renders human avatar movements from dynamic camera videos. AMG is the first method that enables multi-person diffusion video generation with precise control over camera positions, human motions, and background style. We also demonstrate through extensive evaluation that it outperforms existing human video generation methods conditioned on pose sequences or driving videos in terms of realism and adaptability.

📄 PDF Abstract BibTeX arXiv:2409.01502

Code (1)

zshyang/amg 공식 구현 pytorch

Tasks

Video Generation

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음
Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

MagicAvatar: Multimodal Avatar Generation and Animation

2023-08-28 · Jianfeng Zhang, Hanshu Yan, Zhongcong Xu, Jiashi Feng 외

This report presents MagicAvatar, a framework for multimodal video generation and animation of human avatars. Unlike most existing methods that generate avatar-centric videos directly from multimodal inputs (e.g., text p…

Video Generation

InstructAvatar: Text-Guided Emotion and Motion Control for Avatar Generation

2024-05-24 · Yuchi Wang, Junliang Guo, Jianhong Bai, Runyi Yu 외

Recent talking avatar generation models have made strides in achieving realistic and accurate lip synchronization with the audio, but often fall short in controlling and conveying detailed expressions and emotions of the…

DexAvatar: 3D Sign Language Reconstruction with Hand and Body Pose Priors

2025-12-24 · Kaustubh Kundu, Hrishav Bakul Barua, Lucy Robertson-Bell, Zhixi Cai 외 arxiv

The trend in sign language generation is centered around data-driven generative methods that require vast amounts of precise 2D and 3D human pose data to achieve an acceptable generation quality. However, currently, most…

3D Human Pose Estimation

Making Avatars Interact: Towards Text-Driven Human-Object Interaction for Controllable Talking Avatars

2026-02-02 · Youliang Zhang, Zhengguang Zhou, Zhentao Yu, Ziyao Huang 외 arxiv

Generating talking avatars is a fundamental task in video generation. Although existing methods can generate full-body talking avatars with simple human motion, extending this task to grounded human-object interaction (G…

Video Generation

Kling-Avatar: Grounding Multimodal Instructions for Cascaded Long-Duration Avatar Animation Synthesis

2025-09-11 · Yikang Ding, Jiwen Liu, Wenyuan Zhang, Zekun Wang 외 arxiv

Recent advances in audio-driven avatar video generation have significantly enhanced audio-visual realism. However, existing methods treat instruction conditioning merely as low-level tracking driven by acoustic or visual…

Domain GeneralizationVideo Generation