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

• URL: http://arxiv.org/abs/2506.11924v2
• Date: Thu, 26 Jun 2025 15:26:54 GMT
• Title: Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation
• Authors: Min-Seop Kwak, Junho Kim, Sangdoo Yun, Dongyoon Han, Taekyoung Kim, Seungryong Kim, Jin-Hwa Kim,
• Abstract summary: We introduce a diffusion-based framework that performs aligned novel view image and geometry generation via a warping-and-inpainting methodology.<n>Method leverages off-the-shelf geometry predictors to predict partial geometries viewed from reference images.<n>Cross-modal attention distillation is proposed to ensure accurate alignment between generated images and geometry.
• Score: 62.87088388345378
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: 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.

Related papers

• DreamPolish: Domain Score Distillation With Progressive Geometry Generation [66.94803919328815]
  We introduce DreamPolish, a text-to-3D generation model that excels in producing refined geometry and high-quality textures. In the geometry construction phase, our approach leverages multiple neural representations to enhance the stability of the synthesis process. In the texture generation phase, we introduce a novel score distillation objective, namely domain score distillation (DSD), to guide neural representations toward such a domain.
  arXiv  Detail & Related papers  (2024-11-03T15:15:01Z)
• Spurfies: Sparse Surface Reconstruction using Local Geometry Priors [8.260048622127913]
  We introduce Spurfies, a novel method for sparse-view surface reconstruction. It disentangles appearance and geometry information to utilize local geometry priors trained on synthetic data. We validate our method on the DTU dataset and demonstrate that it outperforms previous state of the art by 35% in surface quality.
  arXiv  Detail & Related papers  (2024-08-29T14:02:47Z)
• G-NeRF: Geometry-enhanced Novel View Synthesis from Single-View Images [45.66479596827045]
  We propose a Geometry-enhanced NeRF (G-NeRF), which seeks to enhance the geometry priors by a geometry-guided multi-view synthesis approach. To tackle the absence of multi-view supervision for single-view images, we design the depth-aware training approach.
  arXiv  Detail & Related papers  (2024-04-11T04:58:18Z)
• Towards Geometric-Photometric Joint Alignment for Facial Mesh Registration [3.1932242398896964]
  This paper presents a Geometric-Photometric Joint Alignment(GPJA) method.<n>It aligns discrete human expressions at pixel-level accuracy by combining geometric and photometric information.<n>This consistency benefits face animation, re-parametrization, and other batch operations for face modeling and applications with enhanced efficiency.
  arXiv  Detail & Related papers  (2024-03-05T03:39:23Z)
• Adaptive Surface Normal Constraint for Geometric Estimation from Monocular Images [56.86175251327466]
  We introduce a novel approach to learn geometries such as depth and surface normal from images while incorporating geometric context. Our approach extracts geometric context that encodes the geometric variations present in the input image and correlates depth estimation with geometric constraints. Our method unifies depth and surface normal estimations within a cohesive framework, which enables the generation of high-quality 3D geometry from images.
  arXiv  Detail & Related papers  (2024-02-08T17:57:59Z)
• Explicit Correspondence Matching for Generalizable Neural Radiance Fields [49.49773108695526]
  We present a new NeRF method that is able to generalize to new unseen scenarios and perform novel view synthesis with as few as two source views. The explicit correspondence matching is quantified with the cosine similarity between image features sampled at the 2D projections of a 3D point on different views. Our method achieves state-of-the-art results on different evaluation settings, with the experiments showing a strong correlation between our learned cosine feature similarity and volume density.
  arXiv  Detail & Related papers  (2023-04-24T17:46:01Z)
• Geometry-biased Transformers for Novel View Synthesis [36.11342728319563]
  We tackle the task of synthesizing novel views of an object given a few input images and associated camera viewpoints. Our work is inspired by recent 'geometry-free' approaches where multi-view images are encoded as a (global) set-latent representation. We propose 'Geometry-biased Transformers' (GBTs) that incorporate geometric inductive biases in the set-latent representation-based inference.
  arXiv  Detail & Related papers  (2023-01-11T18:59:56Z)
• PVSeRF: Joint Pixel-, Voxel- and Surface-Aligned Radiance Field for Single-Image Novel View Synthesis [52.546998369121354]
  We present PVSeRF, a learning framework that reconstructs neural radiance fields from single-view RGB images. We propose to incorporate explicit geometry reasoning and combine it with pixel-aligned features for radiance field prediction. We show that the introduction of such geometry-aware features helps to achieve a better disentanglement between appearance and geometry.
  arXiv  Detail & Related papers  (2022-02-10T07:39:47Z)
• Deep Geometric Texture Synthesis [83.9404865744028]
  We propose a novel framework for synthesizing geometric textures. It learns texture statistics from local neighborhoods of a single reference 3D model. Our network displaces mesh vertices in any direction, enabling synthesis of geometric textures.
  arXiv  Detail & Related papers  (2020-06-30T19:36:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.