Shadow determination and compensation for face recognition

Li, Yi; Liu, Xiaodong; Gao, Zhijun

doi:10.1007/s13042-013-0207-z

Cited by 10 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The desired number would be an average intensity value around the middle, which indicates that the histogram is uniform, covering different pixel values from both tones. Third, the clarity measure proposed by Li et al [55] is used according to the following equation

C = \frac{A}{E_{n}}

where A corresponds to the average intensity number and E n is the number of edge pixels extracted by a canny operator. A lower clarity value means more edges and, as a consequence, more shadows.…”

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

“…Then, a post‐processing step is performed to reduce noise and highlight facial texture features. Li et al [55] introduced the concepts of average intensity, kurtosis and clarity to adaptively determine images with shadows in a training set and move them to a shadow set. Kurtosis measures how outlier‐prone the distribution of face images is; on the other hand, clarity is obtained by dividing the average intensity by the number of edge pixels in the image.…”

Section: Unlightingmentioning

confidence: 99%

See 2 more Smart Citations

Addressing the illumination challenge in two‐dimensional face recognition: a survey

Ochoa-Villegas

Nolazco-Flores

Barron-Cano

et al. 2015

IET Computer Vision

View full text Add to dashboard Cite

Uncontrolled illumination is one of the most widely researched and most encountered face recognition challenges in recent years. In this study, the authors propose the division of algorithms into two categories: (i) relighting and (ii) unlighting. Relighting methods try to match the probe's illumination conditions using a subset of representative gallery images, while unlighting methods seek to suppress the variations. A total of 64 state-of-the-art methods are summarised and categorised in each of the groups. To make this work concise and easy to follow, they restricted themselves to selected conferences/journals and they limited the number of approaches reviewed. Also, eight past state-of-the-art approaches are used in both identification and verification experiments. However, only significant reported results from all methods were compared and organised in tables. The author's main objective is not to provide an exhaustive analysis of each category, but to present a collection of papers that can be useful in identifying research directions. Results indicate that unlighting methods are a better and a practical solution to address illumination challenges.

show abstract