Learning by a generation approach to appearance-based object recognition
Learning by a generation approach to appearance-based object recognition
复制标题
通过基于外观的对象识别的生成方法进行学习
DOI：
10.1109/icpr.1996.545985
复制
发表时间：
1996
期刊：
影响因子：
--
通讯作者：
S. Nayar
中科院分区：
文献类型：
--
作者：
H. Murase;S. Nayar
我们提出了一种方法，用于生成学习样本的外观为基础的对象识别。在许多实际情况下，获得大量的学习样本并不容易。所提出的方法学习对象模型从大量的生成的样本来自少量的实际观察到的图像。该学习算法有两个步骤：1）通过图像插值或图像变形生成大量图像，以及2）使用参数特征空间表示压缩大样本集。我们比较了我们的方法与以前的方法，插值样本点的特征空间，并显示我们的方法的性能是上级。对4个目标的432幅图像样本进行了实验，验证了该方法的有效性。
We propose a methodology for the generation of learning samples in appearance-based object recognition. In many practical situations, it is not easy to obtain a large number of learning samples. The proposed method learns object models from a large number of generated samples derived from a small number of actually observed images. The learning algorithm has two steps: 1) generation of a large number of images by image interpolation, or image deformation, and 2) compression of the large sample sets using parametric eigenspace representation. We compare our method with the previous methods that interpolate sample points in eigenspace, and show the performance of our method to be superior. Experiments were conducted for 432 image samples for 4 objects to demonstrate the effectiveness of the method.