Dissimilarity Representations for Low-Resolution Face Recognition

Hernández-Durán, Mairelys; Cheplygina, Veronika; Plasencia-Calaña, Yenisel

doi:10.1007/978-3-319-24261-3_6

Cited by 4 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2015, Hernandez et al [31] argued that using the dissimilarity representation was better than utilizing traditional feature representation and they yield an experimental conclusion that when training images are down-scaled and then up-scaled while the test images are up-scaled, the multidimensional matching performance was the best approach to use. A system based on improving local phase equalization [116] was proposed by Xiao et al They improved the original LPQ in several aspects: First, they extend LPQ to "LPQ plus" by directly quantizing the Fourier transformation of the blurry image densely.…”

Section: Methodsmentioning

confidence: 99%

Face Recognition in Low Quality Images: A Survey

Li¹,

Prieto²,

Mery³

et al. 2018

Preprint

View full text Add to dashboard Cite

Low-quality face recognition (LQFR) has received increasing attention over the past few years. There are numerous potential uses for systems with LQFR capability in real-world environments when high-resolution or high-quality images are difficult or impossible to capture. One of the significant potential application domains for LQFR systems is video surveillance. As the number of surveillance cameras increases (especially in urban environments), the videos that they capture will need to be processed automatically. However, those videos are usually captured with large standoffs, challenging illumination conditions, and diverse angles of view. Faces in these images are generally small in size. Past work on this topic has employed techniques such as super-resolution processing, deblurring, or learning a relationship between different resolution domains. In this paper, we provide a comprehensive review of approaches to low-quality face recognition in the past six years. First, a general problem definition is given, followed by a systematic analysis of the works on this topic by category. Next, we highlight the relevant data sets and summarize their respective experimental results. Finally, we discuss the general limitations and propose priorities for future research.

show abstract

Section: Methodsmentioning

confidence: 99%