PulseAugur
实时 10:08:12

New PanoSeeker Agent Tackles Active Panoramic Referring Segmentation

研究人员提出了一项名为主动全景指代表达分割(Active Panoramic Referring Segmentation, APRS)的新任务,以解决当前分割模型在动态、360度环境中的局限性。他们提出了PanoSeeker,一个利用视觉语言模型(Vision-Language Model)和空间视觉记忆EgoSphere的智能体,以在连续的360度空间中高效地搜索和分割物体。PanoSeeker将连续的观测整合到一个统一的表示中,以规划最优搜索轨迹,在一个新创建的APRS基准测试中表现优于现有方法。 AI

影响 引入了新的任务和智能体,用于具身人工智能(embodied AI),有望改善现实世界机器人应用中的物体交互和分割。

排序理由 学术论文,详细介绍了新任务和提出的模型。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

New PanoSeeker Agent Tackles Active Panoramic Referring Segmentation

报道来源 [2]

  1. arXiv cs.CV TIER_1 English(EN) · Song Tang, Shuming Hu, Xincheng Shuai, Henghui Ding, Yu-Gang Jiang ·

    Seek to Segment: Active Perception for Panoramic Referring Segmentation

    arXiv:2607.02497v1 Announce Type: new Abstract: Existing referring segmentation models passively process static images captured from fixed perspectives, limiting their applicability in Embodied AI, where agents must perform active perception in the continuous 360$^\circ$ environm…

  2. arXiv cs.CV TIER_1 English(EN) · Yu-Gang Jiang ·

    Seek to Segment: Active Perception for Panoramic Referring Segmentation

    Existing referring segmentation models passively process static images captured from fixed perspectives, limiting their applicability in Embodied AI, where agents must perform active perception in the continuous 360$^\circ$ environments. To bridge this gap, we introduce a novel t…