---
title: "Coarse3D: Controllable and Faithful Sparse Structure Generation for Image-to-3D"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.41231v1"
permalink: "https://feed.astralor.com/items/itm_afc41ed7d1f4f7f83d998927"
published: "2026-10-30T04:09:19.000Z"
score: 54
channel: "ai"
tags: ["推理效率","Coarse3D","Controllable","Faithful Sparse Structure","Generation","Image-to-3D Coarse-to-fine","However","Coarse-to-fine"]
---

# Coarse3D: Controllable and Faithful Sparse Structure Generation for Image-to-3D

## 速览

Coarse-to-fine 3D-native generation relies heavily on the first-stage sparse structure, which determines the global orientation and overall geometry for subsequent refinement. However, existing coarse generators suffer …

## 关键点

- Coarse-to-fine 3D-native generation relies heavily on the first-stage sparse structure, which determines the global orientation and overall geometry for subsequent refinement.
- However, existing coarse generators suffer from two key limitations: their orientation is difficult to control due to ambiguous semantic canonicalization, and their structures are often inconsistent with input images du…
- We propose Coarse3D, a sparse-structure generation model that produces orientation-controllable and structurally faithful voxel geometry from arbitrary unposed images.

## 全文

Coarse-to-fine 3D-native generation relies heavily on the first-stage sparse structure, which determines the global orientation and overall geometry for subsequent refinement. However, existing coarse generators suffer from two key limitations: their orientation is difficult to control due to ambiguous semantic canonicalization, and their structures are often inconsistent with input images due to weak 2D-3D grounding. We propose Coarse3D, a sparse-structure generation model that produces orientation-controllable and structurally faithful voxel geometry from arbitrary unposed images. Coarse3D introduces three designs: a first-frame anchored canonical space that aligns generation with the first input view while preserving object-centric normalization; a dual-tower DiT that couples generation with auxiliary feed-forward reconstruction to learn spatially grounded features; and spatial-prior guided attention that regularizes cross-attention with pixel-to-voxel correspondence. Coarse3D can directly replace the coarse stage of existing pipelines. Experiments show that it substantially improves sparse structure quality and consistently enhances final 3D generation, achieving state-of-the-art performance.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_afc41ed7d1f4f7f83d998927