CrossTimeEdit: A Decade-Spanning Cross-View Dataset and Reward-Guided Editing for Historical Street-View Generation
- 类型：arxiv
- 标识：2609.36616
- 链接：https://arxiv.org/abs/2609.36616
- 主分类：llm-infra
- 形态：method
- TLDR：Historical street-view imagery records urban evolution, but uneven coverage leaves substantial gaps in historical records. Generating plausible past appearances requires restoring changed structures while preserving persistent scene content. We construct VIGOR-his, a decade-spanning cross-view dataset containing 43,653 location-level quadruplets across 11 cities on three continents. Its automated pipeline performs spatial pairing, consistency screening, change classification, and the generation and validation of satellite-based change descriptions and local editing instructions. Based on VIGOR
- 待LLM分类：否
- 标题中文：CrossTimeEdit：跨十年的跨视图数据集与面向历史街景生成的奖励引导编辑
- TLDR中文：历史街景图像记录了城市演变，但覆盖不均使历史记录存在大量空白。生成合理的过去外观需要恢复已变化的结构，同时保留持续存在的场景内容。我们构建了 VIGOR-his，一个跨十年的跨视图数据集，包含 11 个城市、3 个大洲的 43,653 个位置级四元组。其自动化流水线执行空间配对、一致性筛选、变化分类，以及基于卫星的变化描述与局部编辑指令的生成与验证。基于 VIGOR
- 来源文件：
- /inbox/tom/_candidates/2026-10-13-agent-rag-longcontext-candidates.json