Discriminative markings are often highly localized, leading traditional object recognition approaches to struggle with the large pose variation often
present in these domains.

As described in [23], what often differentiates basic-level categories is the presence or absence of parts (e.g. an elephant has 4 legs and a trunk), whereas subordinate
categories are more often discriminated by subtle variations in the shape, size and/or appearance properties of these parts (e.g. elephant species can be distinguished by
localized cues such as ear shape and size).

While the limited supervision required for the DPM is advantageous, the latent parts provide no semantic information about the object which makes
pose-normalization challenging.

Our goal is to use DPM to localize the parts and pool the pose-normalized image features induced by the part locations.

The underlying principle in pose-normalization is that one can decompose an object’s appearance as observed in one image and compare it to the same object (or object category)
as observed in a different image.

总结：dpm的作用为把部件当作proposal，然后提取核特征，如梯度，色彩等，进行分类。而服装的是利用部件的位置不同进行分类。因此，使用的方法是根据不同的形状确定部件之间的相对位置关系。直接进行分类。