Learning hierarchical models of scenes, objects, and parts
Learning hierarchical models of scenes, objects, and parts
复制标题
DOI：
10.1109/iccv.2005.137
复制
发表时间：
2005-10
期刊：
影响因子：
--
通讯作者：
Erik B. Sudderth;A. Torralba;W. Freeman;A. Willsky
中科院分区：
文献类型：
--
作者：
Erik B. Sudderth;A. Torralba;W. Freeman;A. Willsky
我们描述了一个分层的概率模型，用于在杂乱的自然场景中检测和识别物体。该模型是基于一组部分，描述了预期的外观和位置，在一个对象为中心的坐标系中，由一个低层次的兴趣运营商检测到的功能。每个对象类别在这些部分上都有自己的分布，这些部分在对象之间共享。我们通过Gibbs采样器学习该模型的参数，该采样器使用图形模型的结构来分析许多参数的平均值。应用于孤立对象的图像数据库，对象之间的部分共享提高了检测精度时，很少的训练样本。我们也将此阶层架构延伸至包含多个物件的场景
We describe a hierarchical probabilistic model for the detection and recognition of objects in cluttered, natural scenes. The model is based on a set of parts which describe the expected appearance and position, in an object centered coordinate frame, of features detected by a low-level interest operator. Each object category then has its own distribution over these parts, which are shared between objects. We learn the parameters of this model via a Gibbs sampler which uses the graphical model's structure to analytically average over many parameters. Applied to a database of images of isolated objects, the sharing of parts among objects improves detection accuracy when few training examples are available. We also extend this hierarchical framework to scenes containing multiple objects