RGB-D object discovery via multi-scene analysis
RGB-D object discovery via multi-scene analysis
复制标题
DOI：
10.1109/iros.2011.6095116
复制
发表时间：
2011-12
期刊：
影响因子：
--
通讯作者：
E. Herbst;Xiaofeng Ren;D. Fox
中科院分区：
文献类型：
--
作者：
E. Herbst;Xiaofeng Ren;D. Fox
我们引入了一种从RGB-D（颜色加深）数据中发现对象发现的算法，这是基于使用RGB-D摄像机进行3-D重建的最新进展。一组3-D地图是从多次访问到同一场景的。我们引入了一个多场景MRF模型，以检测在访问，形状，可见性和颜色提示之间移动的对象。我们使用二-D和3-D匹配的候选对象之间的相似性，并应用光谱聚类从嘈杂的链接中推断对象簇。即使对象无纹理或与其他对象具有相同的形状，我们的方法也可以牢固地检测对象及其在场景之间的运动。
We introduce an algorithm for object discovery from RGB-D (color plus depth) data, building on recent progress in using RGB-D cameras for 3-D reconstruction. A set of 3-D maps are built from multiple visits to the same scene. We introduce a multi-scene MRF model to detect objects that moved between visits, combining shape, visibility, and color cues. We measure similarities between candidate objects using both 2-D and 3-D matching, and apply spectral clustering to infer object clusters from noisy links. Our approach can robustly detect objects and their motion between scenes even when objects are textureless or have the same shape as other objects.