{"slug": "axolotl3d-a-unified-framework-for-faithful-3d-shape-completion", "title": "Axolotl3D: a Unified Framework for Faithful 3D Shape Completion", "summary": "Axolotl3D, a multi-modal and occlusion-aware 3D generation model introduced in arXiv:2607.20660v1, jointly conditions on images, visibility masks, camera parameters, and a partial point cloud to enable faithful 3D shape completion under diverse conditioning signals. The model achieves state-of-the-art performance on Toys4K and OmniObject3D in both clean and occluded settings, as well as strong results in real-world reconstruction and geometry-consistent editing.", "body_md": "arXiv:2607.20660v1 Announce Type: new\nAbstract: Recent 3D generative models produce high-quality geometry from a single image using large-scale priors and diffusion architectures. However, they assume complete visibility and single-view inputs, limiting applicability in multi-view, occluded, or editing scenarios. Although prior works address these challenges individually, they lack a unified framework for controllable 3D completion under diverse conditioning signals. We present Axolotl3D, a multi-modal and occlusion-aware 3D generation model that jointly conditions on images, visibility masks, camera parameters, and a partial point cloud. The point cloud serves as a geometric anchor promoting faithful shape completion, while camera parameters ensure consistent multi-view alignment in a shared 3D coordinate system. A unified training strategy synthesizes diverse conditioning regimes from large-scale 3D data, enabling robust cross-modal reasoning. Experiments on Toys4K and OmniObject3D demonstrate state-of-the-art performance under both clean and occluded settings, as well as strong results in real-world reconstruction and geometry-consistent editing.", "url": "https://wpnews.pro/news/axolotl3d-a-unified-framework-for-faithful-3d-shape-completion", "canonical_source": "https://arxiv.org/abs/2607.20660", "published_at": "2026-07-24 04:00:00+00:00", "updated_at": "2026-07-24 04:27:57.725301+00:00", "lang": "en", "topics": ["computer-vision", "generative-ai", "machine-learning"], "entities": ["Axolotl3D", "Toys4K", "OmniObject3D"], "alternates": {"html": "https://wpnews.pro/news/axolotl3d-a-unified-framework-for-faithful-3d-shape-completion", "markdown": "https://wpnews.pro/news/axolotl3d-a-unified-framework-for-faithful-3d-shape-completion.md", "text": "https://wpnews.pro/news/axolotl3d-a-unified-framework-for-faithful-3d-shape-completion.txt", "jsonld": "https://wpnews.pro/news/axolotl3d-a-unified-framework-for-faithful-3d-shape-completion.jsonld"}}