CoTDet: 任务驱动目标检测的可用性知识提示
CoTDet: Affordance Knowledge Prompting for Task Driven Object Detection
AI总结：
针对任务驱动目标检测中类别多样性的挑战，本文提出探索基本可用性而非类别，利用多级思维链提示从大语言模型提取可用性知识，并构建知识条件检测框架CoTDet，显著提升检测性能。
摘要：
任务驱动目标检测旨在检测图像中适合执行任务的物体实例。其挑战在于，可用于任务的目标类别过于多样，无法局限于传统目标检测的封闭目标词汇集。简单地将常见物体的类别和视觉特征映射到任务上无法应对这一挑战。在本文中，我们提出探索基本可用性而非目标类别，即使不同物体能够完成同一任务的共同属性。此外，我们提出了一种新颖的多级思维链提示（MLCoT），从大型语言模型中提取可用性知识，其中包含从任务到物体示例再到带有理由的基本视觉属性的多级推理步骤。进一步地，为了充分利用知识以利于目标识别和定位，我们提出了一种知识条件检测框架，即CoTDet。它使检测器根据知识生成目标查询并回归边界框。实验结果表明，我们的CoTDet一致且显著地优于最先进的方法（+15.6 box AP和+14.8 mask AP），并能生成物体为何被检测为可执行任务的理由。
英文摘要：
Task driven object detection aims to detect object instances suitable for affording a task in an image. Its challenge lies in object categories available for the task being too diverse to be limited to a closed set of object vocabulary for traditional object detection. Simply mapping categories and visual features of common objects to the task cannot address the challenge. In this paper, we propose to explore fundamental affordances rather than object categories, i.e., common attributes that enable different objects to accomplish the same task. Moreover, we propose a novel multi-level chain-of-thought prompting (MLCoT) to extract the affordance knowledge from large language models, which contains multi-level reasoning steps from task to object examples to essential visual attributes with rationales. Furthermore, to fully exploit knowledge to benefit object recognition and localization, we propose a knowledge-conditional detection framework, namely CoTDet. It conditions the detector from the knowledge to generate object queries and regress boxes. Experimental results demonstrate that our CoTDet outperforms state-of-the-art methods consistently and significantly (+15.6 box AP and +14.8 mask AP) and can generate rationales for why objects are detected to afford the task.
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。
