arXiv:2610.11215 · Agent 智能体
GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images
GATOR：基于随手拍图像的生成式与 Agent 式 3D 物体重建
GATOR: Generative and Agentic 3D Object Reconstruction From Casual Images
- 类型：arxiv
- 标识：2610.11215
- 链接：https://arxiv.org/abs/2610.11215
- 主分类：agent
- 形态：method
- 被引：0
- 被引来源：Semantic Scholar
- S2被引：0
- 影响力被引：0
- TLDR：GATOR is presented, a generative and agentic framework that recovers textured object assets and their scene-relative pose from one or more images and achieves strong geometric and appearance fidelity while recovering scene-relative pose from sparse observations.
- 待LLM分类：否
- 标题中文：GATOR：基于随手拍图像的生成式与 Agent 式 3D 物体重建
- TLDR中文：提出 GATOR，一个生成式与 Agent 式框架，能够从一张或多张图像恢复带纹理的物体资产及其相对于场景的位姿，在稀疏观测下恢复场景相对位姿的同时实现较强的几何与外观保真度。
- 来源文件：
- /inbox/tom/_candidates/2026-10-13-rag-retrieval-reranking-candidates.json
- /inbox/tom/_candidates/2026-10-13-agent-rag-longcontext-candidates.json
- [S2 enrich]