BSIF: Binarized statistical image features

By Juho Kannala, Esa Rahtu. 2012

625 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Statistical Meta-Analysis of Presentation Attacks for Secure Multibiometric Systems (2016, positive): “It accounts for illumination variations as in [43], and then computes a BSIF descriptor [44].”
Robust Scheme for Iris Presentation Attack Detection Using Multiscale Binarized Statistical Image Features (2015, methodology): “In this work, we have employed the open-source filters [29] that are trained using 50000 image patches randomly sampled from 13 different natural scenic images [30].”
Past, Present, and Future of Face Recognition: A Review (2020, methodology): “Kannala and Rahtu [95] (2012) proposed a method for building local image descriptors that encode texture information efficiently and are proper for the description of image regions based on histograms.”
A Classification-Selection Approach for Self Updating of Face Verification Systems Under Stringent Storage and Computational Requirements (2015, methodology): “A set of Binarized Statistically Independent Features (BSIFs) is extracted from each block and concatenated in order to form the final feature vector [8].”
Deep Representations for Iris, Face, and Fingerprint Spoofing Detection (2014, methodology): “We follow this approach in this work, assuming little a priori knowledge about acquisition-level biometric spooﬁng and exploring deep representations of the data.”
Face Spoofing Detection Using Colour Texture Analysis (2016, methodology): “4) Binarized Statistical Image Features (BSIF): Similarly to LBP, the BSIF descriptor [53] computes a binary code string for each pixel in an image where each bit is obtained by first convolving the image with a linear filter and then binarizing the filter responses.”
Robust Face Recognition via Multimodal Deep Face Representation (2015, background): “Representative learning-based descriptors include Binarised Statistical Image Features (BSIF) [7], [14] and Discriminant Face Descriptor (DFD) [15], et al. .”
Face Presentation Attack with Latex Masks in Multispectral Videos (2017, methodology): “VIS NIR Thermal HOG [6] 34.9 45.5 24.5 ULBP [15] 38.7 49.0 21.3 BSIF [11] 29.2 43.3 15.0 LPQ [16] 40.0 43.8 13.7 RDWT + Haralick [2] 32.9 42.0 10.8 on frame-wise basis.”
COVID-19 identification in chest X-ray images on flat and hierarchical classification scenarios (2020, unknown): “Proposed by Kannala and Rahtu [48], the BSIF texture descriptor was initially proposed for texture classification particularly on face recognition tasks.”
Local binary features for texture classification: Taxonomy and experimental study (2017, methodology): “Weighted Rotation and Scale Invariant LBP WRSI_LBP Texture Classification 2015 [90] Local Energy Pattern LEP Texture Classification 2013 [84] Binarized Statistical Image Features BSIF Face Recognition 2012 [85] Local Radon Binary Pattern LRBP Face Recognition 2012 [87] Combining multiple LBP type fe”
A Survey on Anti-Spooﬁng Schemes for Fingerprint Recognition Systems (2014, background): “Such a value represents the local descriptor of the image intensity pattern in the neighborhood of the considered pixel [Kannala and Rahtu 2012].”