Binary Feature Descriptors Research Articles

Learning Compact Binary Face Descriptor for Face Recognition.

Binary feature descriptors such as local binary patterns (LBP) and its variations have been widely used in many face recognition systems due to their excellent robustness and strong discriminative power. However, most existing binary face descriptors are hand-crafted, which require strong prior knowledge to engineer them by hand. In this paper, we propose a compact binary face descriptor (CBFD) feature learning method for face representation and recognition. Given each face image, we first extract pixel difference vectors (PDVs) in local patches by computing the difference between each pixel and its neighboring pixels. Then, we learn a feature mapping to project these pixel difference vectors into low-dimensional binary vectors in an unsupervised manner, where 1) the variance of all binary codes in the training set is maximized, 2) the loss between the original real-valued codes and the learned binary codes is minimized, and 3) binary codes evenly distribute at each learned bin, so that the redundancy information in PDVs is removed and compact binary codes are obtained. Lastly, we cluster and pool these binary codes into a histogram feature as the final representation for each face image. Moreover, we propose a coupled CBFD (C-CBFD) method by reducing the modality gap of heterogeneous faces at the feature level to make our method applicable to heterogeneous face recognition. Extensive experimental results on five widely used face datasets show that our methods outperform state-of-the-art face descriptors.

Real-time monocular image-based 6-DoF localization

In this paper we present a new real-time image-based localization method for scenes that have been reconstructed offline using structure from motion. From input video, our method continuously computes six-degree-of-freedom camera pose estimates by efficiently tracking natural features and matching them to 3D points reconstructed by structure from motion. Our main contribution lies in efficiently interleaving a fast keypoint tracker that uses inexpensive binary feature descriptors with a new approach for direct 2D-to-3D matching. Our 2D-to-3D matching scheme avoids the need for online extraction of scale-invariant features. Instead, offline we construct an indexed database containing multiple DAISY descriptors per 3D point extracted at multiple scales. The key to the efficiency of our method is invoking DAISY descriptor extraction and matching sparingly during localization, and in distributing this computation over a temporal window of successive frames. This enables the system to run in real-time and achieve low per-frame latency over long durations. Our algorithm runs at over 30 Hz on a laptop and at 12 Hz on a low-power computer suitable for onboard computation on a mobile robot such as a micro-aerial vehicle. We have evaluated our method using ground truth and present results on several challenging indoor and outdoor sequences.

Learning Image Descriptors with Boosting.

We propose a novel and general framework to learn compact but highly discriminative floating-point and binary local feature descriptors. By leveraging the boosting-trick we first show how to efficiently train a compact floating-point descriptor that is very robust to illumination and viewpoint changes. We then present the main contribution of this paper-a binary extension of the framework that demonstrates the real advantage of our approach and allows us to compress the descriptor even further. Each bit of the resulting binary descriptor, which we call BinBoost, is computed with a boosted binary hash function, and we show how to efficiently optimize the hash functions so that they are complementary, which is key to compactness and robustness. As we do not put any constraints on the weak learner configuration underlying each hash function, our general framework allows us to optimize the sampling patterns of recently proposed hand-crafted descriptors and significantly improve their performance. Moreover, our boosting scheme can easily adapt to new applications and generalize to other types of image data, such as faces, while providing state-of-the-art results at a fraction of the matching time and memory footprint.

SPHORB: A Fast and Robust Binary Feature on the Sphere

In this paper, we propose SPHORB, a new fast and robust binary feature detector and descriptor for spherical panoramic images. In contrast to state-of-the-art spherical features, our approach stems from the geodesic grid, a nearly equal-area hexagonal grid parametrization of the sphere used in climate modeling. It enables us to directly build fine-grained pyramids and construct robust features on the hexagonal spherical grid, thus avoiding the costly computation of spherical harmonics and their associated bandwidth limitation. We further study how to achieve scale and rotation invariance for the proposed SPHORB feature. Extensive experiments show that SPHORB consistently outperforms other existing spherical features in accuracy, efficiency and robustness to camera movements. The superior performance of SPHORB has also been validated by real-world matching tests.

Binary feature from intensity quantization and weakly spatial contextual coding for image search

During the past few years, a number of local binary feature descriptors for images have been proposed, e.g. BRIEF, ORB, BRISK, and FREAK. The binary descriptors have several advantages over the well-established floating vector descriptors such as SIFT and SURF, for their fast computing speed and much low memory consumption. Nevertheless, the binary descriptor still suffer from the poor performance in computer vision applications for its low discriminative power and Hamming distance metric. To improve the capability of binary descriptors, some works focusing on the points selection pattern algorithm in the descriptor extraction are proposed. These works mostly adopt more optimal selection pattern (BRISK and FREAK) to enhance the performance of binary descriptors, rather than random chosen pattern. In our study, however, the points selection algorithm does not have much contributions for the promotion on performance. Therefore, in this paper, we try to solve the problem of low discriminative power and robustness through two novel methods: Intensity Difference Quantization and Weakly Spatial Context Coding. The experimental results on the public datasets show that our method can significantly boost the performance of binary features and highly enhance the retrieval accuracy of the image search system, even though our proposed method increases slightly memory usage and computing efficiency, compared to the original binary descriptors.

Detection of Region-duplication Forgery in Image Based on Key Points' Binary Descriptors

As a simple and efiective way of image tampering, copy-move forgery was often used as a preferred method to forgers. A new method for detecting and localizing this type of malicious tampering is presented in this paper. The method flrst extracts Harris corners as the key points of the image, then extracts the binary feature descriptors of these key points, and searches the most similarity feature descriptor which satisfy a certain condition according to the descriptors, flnally localizes the tampered regions. The experiment results show that the method can efiectively detect this type of tampering for images. Simultaneously, the matching time is faster than these methods which descriptors’ storage type is ∞oating-point, such as SIFT and SURF.

Matching corners using the informative arc

Corners are important features in images because they typically delimit the boundaries of regions or objects. For real‐time applications, it is essential that corners are detected and matched reliably and rapidly. This study presents two related descriptors which are compatible with standard corner detectors and able to be computed and matched at video rate: one encodes the entire region within a corner, whereas the other describes only the region within an object. The advantage of encoding only the region within an object is demonstrated. The noise stability of the descriptors is assessed and compared with that of the popular binary robust independent elementary feature (BRIEF) descriptor, and the matching performances of the descriptors are compared on video sequences from hand‐held cameras and the PETS2012 database. A statistical analysis shows that performance is indistinguishable from BRIEF.

Is Hamming distance only way for matching binary image feature descriptors?

Brute force matching of binary image feature descriptors is conventionally performed using the Hamming distance. This paper assesses the use of alternative metrics in order to see whether they can produce feature correspondences that yield more accurate homography matrices. Two statistical tests, namely ANOVA (Analysis of Variance) and McNemar's test were employed for evaluation. Results show that Jackard-Needham and Dice metrics can display better performance for some descriptors. Yet, these performance differences were not found to be statistically significant.

Receptive fields selection for binary feature description.

Feature description for local image patch is widely used in computer vision. While the conventional way to design local descriptor is based on expert experience and knowledge, learning-based methods for designing local descriptor become more and more popular because of their good performance and data-driven property. This paper proposes a novel data-driven method for designing binary feature descriptor, which we call receptive fields descriptor (RFD). Technically, RFD is constructed by thresholding responses of a set of receptive fields, which are selected from a large number of candidates according to their distinctiveness and correlations in a greedy way. Using two different kinds of receptive fields (namely rectangular pooling area and Gaussian pooling area) for selection, we obtain two binary descriptors RFDR and RFDG .accordingly. Image matching experiments on the well-known patch data set and Oxford data set demonstrate that RFD significantly outperforms the state-of-the-art binary descriptors, and is comparable with the best float-valued descriptors at a fraction of processing time. Finally, experiments on object recognition tasks confirm that both RFDR and RFDG successfully bridge the performance gap between binary descriptors and their floating-point competitors.

A Fast and Robust Binary Feature Descriptor Based on Tuning Sampling Pattern

Based on two sampling patterns’ characteristics of BRISK and FREAK, a new sampling pattern which results in a better descriptor is proposed. Both sampling-point density and size of overlapping receptive fields on sampling pattern have a significantly effect on the performance of descriptor through analyzing both sampling patterns of BRISK and FREAK. Optima of both factors are got from respective tests of describing local area to be more discriminative and enable the sampling pattern to produce a descriptor with better performance. Through a comparing performance test with state-of-the-art descriptors on benchmark datasets, the proposed descriptor in combination with the SURF detector has been proved to be a high-performance one to capture the most discriminative information in the detected salient region.

A Novel Binary Feature Descriptor for Accelerated Robust Matching

Stereo matching between images is at the base of many computer vision applications. Many traditional image features for matching focus on being robust to view-point changes with huge amount of calculation and memory storage, such as Scale Invariant Feature Transform (SIFT). Binary Robust Independent Elementary Features (BRIEF) is an efficient alternative to traditional SIFT-like features, but it is sensitive to rotation and noise. So BRIEF is not suitable for some natural scene with view-point changes. Nowadays, the deployment of computer vision algorithms on mobile devices with low power of computation and memory capacity has even upped the ante: the goal is to make descriptors efficient to extract, more compact while remaining invariant to scale, rotation and noise. In this paper, we propose a novel feature descriptor with binary strings that is very fast to compute and match with good performance of being invariant to scale, rotation and noise to address the current requirement. The binary descriptors will be directly extracted by comparing the responses of Gaussian filter of the sampling patches that effectively avoid being sensitive to noise. Our experiment results illustrate that our algorithm is fast to compute with low memory load and invariant to view-point change, blur change, brightness change, and JPEG compression.