Yipeng Liu

Efficient and Robust Corner Detectors Based on Second-Order Difference of Contour

As one of the most significant local features of image, corner is widely used in many computer vision tasks. Corner detection aims to achieve the highest possible detection accuracy while minimizing the computational complexity. In this letter, we first introduce a new measurement termed as second-order difference of contour (SODC), and then examine its regular distribution, which is found to provide useful information to distinguish corners from noncorners. Based on the SODC distribution characteristics, we propose two novel corner detectors to measure the response of contour points using Manhattan distance and Euclidean distance, respectively. Numerical experiments demonstrate that the Manhattan detector greatly decreases the computational complexity, while the Euclidean detector outperforms the state-of-the-art corner detectors in terms of repeatability and localization error.

3D interest point detection based on geometric measures and sparse refinement

Three dimensional (3D) interest point detection plays a fundamental role in computer vision. In this paper, we introduce a new method for detecting 3D interest points of 3D mesh models based on geometric measures and sparse refinement (GMSR). The key point of our approach is to calculate the 3D saliency measure using two novel geometric measures, which are defined in multi-scale space to effectively distinguish 3D interest points from edges and flat areas. Those points with local maxima of 3D saliency measure are selected as the candidates of 3D interest points. Finally, we utilize an l0 norm based optimization method to refine the candidates of 3D interest points by constraining the number of 3D interest points. Numerical experiments show that the proposed GMSR based 3D interest point detector outperforms current six state-of-the-art methods for different kinds of 3D mesh models.

Iterative block tensor singular value thresholding for extraction of lowrank component of image data

Tensor principal component analysis (TPCA) is a multi-linear extension of principal component analysis which converts a set of correlated measurements into several principal components. In this paper, we propose a new robust TPCA method to extract the principal components of the multi-way data based on tensor singular value decomposition. The tensor is split into a number of blocks of the same size. The low rank component of each block tensor is extracted using iterative tensor singular value thresholding method. The principal components of the multi-way data are the concatenation of all the low rank components of all the block tensors. We give the block tensor incoherence conditions to guarantee the successful decomposition. This factorization has similar optimality properties to that of low rank matrix derived from singular value decomposition. Experimentally, we demonstrate its effectiveness in two applications, including motion separation for surveillance videos and illumination normalization for face images.

Low rank tensor completion for multiway visual data

Tensor completion recovers missing entries of multiway data. Teh missing of entries could often be caused during teh data acquisition and transformation. In dis paper, we provide an overview of recent development in low rank tensor completion for estimating teh missing components of visual data, e. g. , color images and videos. First, we categorize these methods into two groups based on teh different optimization models. One optimizes factors of tensor decompositions wif predefined tensor rank. Teh other iteratively updates teh estimated tensor via minimizing teh tensor rank. Besides, we summarize teh corresponding algorithms to solve those optimization problems in details. Numerical experiments are given to demonstrate teh performance comparison when different methods are applied to color image and video processing.

Robust corner detection using altitude to chord ratio accumulation

As one of the most significant image local features, corner is widely utilized in many computer vision applications. A number of contour-based corner detection algorithms have been proposed over the last decades, among which the chord-to-point distance accumulation (CPDA) corner detector is reported to produce robust performance in corner detection, especially compared with curvature scale-space (CSS) based corner detectors, which are sensitive to local variation and noise on the contour. In this paper, we investigate the CPDA algorithm in terms of its limitations, and then propose the altitude-to-chord ratio accumulation (ACRA) corner detector based on CPDA approach. Altitude-to-chord ratio is insensitive to the selection of chord length compared with chord-to-point distance, which allows us utilize a single chord instead of the three chords used in CPDA algorithm. Besides, we replace the maximum normalization used in CPDA algorithm with the linear normalization to avoid the uneven data projection. Numerical experiments demonstrate that the proposed ACRA corner detection algorithm outperforms the CPDA approach and other seven state-of-the-art methods in terms of the repeatability and localization error evaluation metrics.

Extended smoothlets: An efficient multi-resolution adaptive transform

Abstract As a family of multiresolution adaptive transforms and a generalization of wedgelets, smoothlets are more efficient in representing images with various sharp edges than other “X-lets”. Smoothlets use a horizon function to model an edge and define transition in a fixed axis direction. Furthermore, the conventional elliptical smoothlets (ES) use half of an ellipse to model edges. Various sharp edges of images may not be expressed well by half of an ellipse with transition only in a fixed axis direction. In this paper, we propose extended smoothlets (ExSmoothlets) framework by using more general characteristic functions with adaptive transition directions. Specifically, two methods of ExSmoothlets, the elliptical ExSmoothlets (EES) and homocentric elliptical ExSmoothlets (HEES) are developed and evaluated in the experiments against the ES. The results show that the proposed EES and HEES can effectively improve the image quality in image approximation and denoising in terms of PSNR.

Learning based 3D keypoint detection with local and global attributes in multi-scale space

Over the last few decades various methods have been proposed by researchers to extract 3D keypoints from the surface of 3D mesh models, but most of them are geometric ones, which are not flexible enough for various applications. In this paper, we propose a new 3D keypoint detection method based on multi-scale neural network (MSNN), which is a tiny neural network and can effectively merge multi-scale information to detect 3D keypoints. Traditional end-to-end learning systems usually require large-scale dataset to do training. However, there are not enough 3D data with ground truth of 3D keypoints. To solve this problem, we perform delicate preprocessing, which effectively enhance the performance of the MSNN based approach. Numerical experiments show that the proposed MSNN 3D keypoint detector not only outperforms other six state-of-the-art geometric based methods, but also achieves better performance than a learning-based method using random forest.