Unverified50% confidenceFactExact time
VDAs replace 3DMM with a learned expression space to break through the expressive ceiling of 3DMM.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedC3D（2015年）将 2D 卷积扩展为 3D 卷积在时空立方体上学习特征，I3D 通过膨胀预训练 ImageNet 权重实现迁移学习74% similarUnverifiedTraditional avatar methods mostly rely on 3D morphable models (3DMM) to drive expressions, but 3DMM has limited expressive capability in regions such as the tongue and the interior of the mouth.71% similarUnverifiedOn the novel-view reenactment task, VDAs on the BecomingLit dataset reduces facial LPIPS by 29% and improves JOD by 66% compared with the strongest single-image 3D avatar method.71% similarUnverified3DMM is a classic parametric face representation proposed by Blanz and Vetter in 1999, whose core idea is to apply PCA dimensionality reduction to a large number of 3D face scans.70% similarUnverifiedIncreasing the resolution of visual input is one of the practical paths to alleviating the limitations of VLM visual encoders.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065789API
curl https://kongchang.com/api/v1/knowledge/claims/1065789MCP
get_claim(id=1065789)