SIE3D: Single-Image Expressive 3D Avatar Generation via Semantic Embedding and Perceptual Expression Loss

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Huang, Zhiqi, Cui, Dulongkai, Hu, Jinglu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918465491697664
author Huang, Zhiqi
Cui, Dulongkai
Hu, Jinglu
author_facet Huang, Zhiqi
Cui, Dulongkai
Hu, Jinglu
contents Generating high-fidelity 3D head avatars from a single image is challenging, as current methods lack fine-grained, intuitive control over expressions via text. This paper proposes SIE3D, a framework that generates expressive 3D avatars from a single image and descriptive text. SIE3D fuses identity features from the image with semantic embedding from text through a novel conditioning scheme, enabling detailed control. To ensure generated expressions accurately match the text, it introduces an innovative perceptual expression loss function. This loss uses a pre-trained expression classifier to regularize the generation process, guaranteeing expression accuracy. Extensive experiments show SIE3D significantly improves controllability and realism, outperforming competitive methods in identity preservation and expression fidelity on a single consumer-grade GPU. Project page: https://huang-zhiqi.github.io/SIE3D/
format Preprint
id arxiv_https___arxiv_org_abs_2509_24004
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SIE3D: Single-Image Expressive 3D Avatar Generation via Semantic Embedding and Perceptual Expression Loss
Huang, Zhiqi
Cui, Dulongkai
Hu, Jinglu
Computer Vision and Pattern Recognition
Generating high-fidelity 3D head avatars from a single image is challenging, as current methods lack fine-grained, intuitive control over expressions via text. This paper proposes SIE3D, a framework that generates expressive 3D avatars from a single image and descriptive text. SIE3D fuses identity features from the image with semantic embedding from text through a novel conditioning scheme, enabling detailed control. To ensure generated expressions accurately match the text, it introduces an innovative perceptual expression loss function. This loss uses a pre-trained expression classifier to regularize the generation process, guaranteeing expression accuracy. Extensive experiments show SIE3D significantly improves controllability and realism, outperforming competitive methods in identity preservation and expression fidelity on a single consumer-grade GPU. Project page: https://huang-zhiqi.github.io/SIE3D/
title SIE3D: Single-Image Expressive 3D Avatar Generation via Semantic Embedding and Perceptual Expression Loss
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.24004