Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Kwak, Min-Seop, Kim, Junho, Yun, Sangdoo, Han, Dongyoon, Kim, Taekyung, Kim, Seungryong, Kim, Jin-Hwa
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911426439806976
author Kwak, Min-Seop
Kim, Junho
Yun, Sangdoo
Han, Dongyoon
Kim, Taekyung
Kim, Seungryong
Kim, Jin-Hwa
author_facet Kwak, Min-Seop
Kim, Junho
Yun, Sangdoo
Han, Dongyoon
Kim, Taekyung
Kim, Seungryong
Kim, Jin-Hwa
contents We introduce a diffusion-based framework that performs aligned novel view image and geometry generation via a warping-and-inpainting methodology. Unlike prior methods that require dense posed images or pose-embedded generative models limited to in-domain views, our method leverages off-the-shelf geometry predictors to predict partial geometries viewed from reference images, and formulates novel-view synthesis as an inpainting task for both image and geometry. To ensure accurate alignment between generated images and geometry, we propose cross-modal attention distillation, where attention maps from the image diffusion branch are injected into a parallel geometry diffusion branch during both training and inference. This multi-task approach achieves synergistic effects, facilitating geometrically robust image synthesis as well as well-defined geometry prediction. We further introduce proximity-based mesh conditioning to integrate depth and normal cues, interpolating between point cloud and filtering erroneously predicted geometry from influencing the generation process. Empirically, our method achieves high-fidelity extrapolative view synthesis on both image and geometry across a range of unseen scenes, delivers competitive reconstruction quality under interpolation settings, and produces geometrically aligned colored point clouds for comprehensive 3D completion. Project page is available at https://cvlab-kaist.github.io/MoAI.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11924
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation
Kwak, Min-Seop
Kim, Junho
Yun, Sangdoo
Han, Dongyoon
Kim, Taekyung
Kim, Seungryong
Kim, Jin-Hwa
Computer Vision and Pattern Recognition
We introduce a diffusion-based framework that performs aligned novel view image and geometry generation via a warping-and-inpainting methodology. Unlike prior methods that require dense posed images or pose-embedded generative models limited to in-domain views, our method leverages off-the-shelf geometry predictors to predict partial geometries viewed from reference images, and formulates novel-view synthesis as an inpainting task for both image and geometry. To ensure accurate alignment between generated images and geometry, we propose cross-modal attention distillation, where attention maps from the image diffusion branch are injected into a parallel geometry diffusion branch during both training and inference. This multi-task approach achieves synergistic effects, facilitating geometrically robust image synthesis as well as well-defined geometry prediction. We further introduce proximity-based mesh conditioning to integrate depth and normal cues, interpolating between point cloud and filtering erroneously predicted geometry from influencing the generation process. Empirically, our method achieves high-fidelity extrapolative view synthesis on both image and geometry across a range of unseen scenes, delivers competitive reconstruction quality under interpolation settings, and produces geometrically aligned colored point clouds for comprehensive 3D completion. Project page is available at https://cvlab-kaist.github.io/MoAI.
title Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.11924