GeCo: Evaluating Geometric Consistency for Video Generation via Motion and Structure

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Gu, Leslie, Hur, Junhwa, Herrmann, Charles, Zhan, Fangneng, Zickler, Todd, Sun, Deqing, Pfister, Hanspeter
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915950004011008
author Gu, Leslie
Hur, Junhwa
Herrmann, Charles
Zhan, Fangneng
Zickler, Todd
Sun, Deqing
Pfister, Hanspeter
author_facet Gu, Leslie
Hur, Junhwa
Herrmann, Charles
Zhan, Fangneng
Zickler, Todd
Sun, Deqing
Pfister, Hanspeter
contents We introduce GeCo, a geometry-grounded metric for jointly detecting geometric deformation and occlusion-inconsistency artifacts in static scenes. By fusing residual motion and depth priors, GeCo produces interpretable, dense consistency maps that reveal these artifacts. We use GeCo to systematically benchmark recent video generation models, uncovering common failure modes, and further employ it as a training-free guidance loss to reduce deformation artifacts during video generation.
format Preprint
id arxiv_https___arxiv_org_abs_2512_22274
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GeCo: Evaluating Geometric Consistency for Video Generation via Motion and Structure
Gu, Leslie
Hur, Junhwa
Herrmann, Charles
Zhan, Fangneng
Zickler, Todd
Sun, Deqing
Pfister, Hanspeter
Computer Vision and Pattern Recognition
We introduce GeCo, a geometry-grounded metric for jointly detecting geometric deformation and occlusion-inconsistency artifacts in static scenes. By fusing residual motion and depth priors, GeCo produces interpretable, dense consistency maps that reveal these artifacts. We use GeCo to systematically benchmark recent video generation models, uncovering common failure modes, and further employ it as a training-free guidance loss to reduce deformation artifacts during video generation.
title GeCo: Evaluating Geometric Consistency for Video Generation via Motion and Structure
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.22274