Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions

Abstract

Abstract. Recognition in uncontrolled situations is one of the most important bottlenecks for practical face recognition systems. We address this by combining the strengths of robust illumination normalization, local texture based face representations and distance transform based matching metrics. Specifically, we make three main contributions: (i) we present a simple and efficient preprocessing chain that eliminates most of the effects of changing illumination while still preserving the essential appearance details that are needed for recognition; (ii) we introduce Local Ternary Patterns (LTP), a generalization of the Local Binary Pattern (LBP) local texture descriptor that is more discriminant and less sensitive to noise in uniform regions; and (iii) we show that replacing local histogramming with a local distance transform based similarity metric further improves the performance of LBP/LTP based face recognition. The resulting method gives state-of-the-art performance on three popular datasets chosen to test recognition under difficult illumination conditions: Face Recognition Grand Challenge version 1 experiment 4, Extended Yale-B, and CMU PIE.

A deep analysis on age estimation

“…Similarly, we have employed an early-fusion strategy, combining the features from the very beginning, before the classification and decision take place. Other strategies could have been used, such as a latefusion strategy, where each feature is coupled with its own classification, and the fusion is performed at decision level, as in Tan and Triggs (2010).…”

Protein classification using texture descriptors extracted from the protein backbone image

“…In particular, we examine variants of the well-known local binary patterns. Our experiments show that the best performance is obtained when the idea of dominant local binary patterns (DLB) (Liao et al, 2009) is combined with local ternary patterns (LTP) (Tan & Triggs, 2007). With DLP, the most frequent rotation invariant patterns are selected.…”

“…To make LBP more robust to noise, (Tan & Triggs, 2007) used Local Ternary Patterns (LTP). In LTP the difference d between x and its neighborhood u is encoded by 3 values according to a threshold τ (here τ=3): 1 if u ≥ x + τ ; -1 if u ≤ x -τ ; else 0.…”

Fusing Gabor and LBP Feature Sets for Kernel-Based Face Recognition

“…Then they are photometrically normalized using the following sequence of steps: strong gamma compression; Difference of Gaussian (DoG) filtering; robust normalization of the range of output variations; and sigmoid-function based compression of any remaining signal highlights. A detailed description of this simple but very effective normalization procedure can be found in [27]. Some examples of preprocessed images are shown in Fig.…”