EAGLE-360: 360°环境中的具身主动全局到局部探索
EAGLE-360: Embodied Active Global-to-Local Exploration in 360$^\circ$
AI总结：
针对多模态大模型在360°全景主动视觉搜索中因极地畸变和连续拓扑建模不足导致的效率低下问题，提出EAGLE-360框架，通过全局先验引导局部探索，结合RoPE Rolling位置编码和GRPO训练，实现近8倍精度提升。
摘要：
尽管多模态大语言模型（MLLMs）在标准视觉理解中展现出卓越能力，但将其应用于360°全景环境中的主动视觉搜索时暴露了根本性局限。具体而言，标准MLLMs难以有效建模全景的固有属性，如严重的极地畸变和连续圆柱拓扑，这显著降低了目标检测精度。因此，现有全景搜索方法试图通过严重依赖碎片化的局部视角来补偿。受限于僵化的初始化和缺乏全局全景先验，这些方法存在短视、低效的探索问题，并在目标移出视野时难以进行鲁棒的错误恢复。为克服这些挑战，我们提出EAGLE-360，一种新颖的具身主动全局到局部探索框架。EAGLE-360并非执行穷举局部搜索，而是利用全局先验建立初始整体视角，迭代推理并逐步缩小搜索空间。在架构上，我们采用RoPE Rolling（一种坐标移位位置编码机制）来无缝建模全景的连续拓扑。为促进这一范式，我们构建了大规模EAGLE-360数据集，包含14000+张4K全景图和70000+轮高质量VQA对话。通过采用集成监督微调（SFT）与组相对策略优化（GRPO）的训练流程，我们有效激发了复杂的空间推理和工具调用能力。大量实验表明，EAGLE-360在360°视觉搜索中建立了新的最先进水平，相比基础模型实现了近8倍的精度提升，同时显著增强了探索效率。
英文摘要：
While Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in standard visual understanding, adapting them for active visual search in 360$^\circ$ panoramic environments exposes fundamental limitations. Specifically, standard MLLMs struggle to effectively model inherent panoramic properties, such as severe polar distortion and continuous cylindrical topologies, which significantly degrades target detection accuracy. Consequently, existing panoramic search methods attempt to compensate by relying heavily on fragmented local viewpoints. Burdened by rigid initialization and a lack of global panoramic priors, these approaches suffer from myopic, inefficient exploration and struggle with robust error recovery when targets are out of view. To overcome these challenges, we propose EAGLE-360, a novel Embodied Active Global-to-Local Exploration framework. Rather than performing exhaustive local searches, EAGLE-360 leverages global priors to establish an initial holistic perspective, iteratively reasoning and progressively narrowing the search space. Architecturally, we adapt RoPE Rolling, a coordinate-shifting positional encoding mechanism, to seamlessly model the continuous topologies of panoramas. To facilitate this paradigm, we construct the large-scale EAGLE-360 dataset, comprising 14,000+ 4K panoramas and 70,000+ rounds of high-quality VQA dialogues. By employing a training pipeline that integrates Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO), we effectively elicit complex spatial reasoning and tool-calling capabilities. Extensive experiments demonstrate that EAGLE-360 establishes a new state-of-the-art for 360$^\circ$ visual search, achieving nearly an 8-fold increase in accuracy over the base model while significantly enhancing exploration efficiency.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
