WLD: A Robust Local Image Descriptor

TLDR

Weber's Law suggests that human perception of a pattern depends on both stimulus change and its original intensity, motivating the development of the robust local descriptor WLD. The paper proposes the Weber Local Descriptor (WLD), a simple yet powerful local image descriptor. WLD is computed by combining a differential excitation term—ratio of a pixel’s intensity differences with its neighbors to its own intensity—and an orientation term given by the pixel’s gradient orientation, then forming a concatenated histogram. Experiments on texture databases Brodatz and KTH‑TIPS2‑a demonstrate that WLD outperforms Gabor and SIFT, while face‑detection tests on MIT+CMU, AR, and CMU profile datasets show performance comparable to state‑of‑the‑art methods.

Abstract

Inspired by Weber's Law, this paper proposes a simple, yet very powerful and robust local descriptor, called the Weber Local Descriptor (WLD). It is based on the fact that human perception of a pattern depends not only on the change of a stimulus (such as sound, lighting) but also on the original intensity of the stimulus. Specifically, WLD consists of two components: differential excitation and orientation. The differential excitation component is a function of the ratio between two terms: One is the relative intensity differences of a current pixel against its neighbors, the other is the intensity of the current pixel. The orientation component is the gradient orientation of the current pixel. For a given image, we use the two components to construct a concatenated WLD histogram. Experimental results on the Brodatz and KTH-TIPS2-a texture databases show that WLD impressively outperforms the other widely used descriptors (e.g., Gabor and SIFT). In addition, experimental results on human face detection also show a promising performance comparable to the best known results on the MIT+CMU frontal face test set, the AR face data set, and the CMU profile test set.