This paper proposes new low-dimensional image features that enable images to be very efficiently matched. Image matching is one of the key technologies for many vision-based applications, including template matching, block motion estimation, video compression, stereo vision, image/video near-duplicate detection, similarity join for image/video database, and so on. Normalized cross correlation (...

Many features of dimensional reduction schemes are determined by the breaking of higher dimensional general covariance associated with the selection of a particular subset of coordinates. By investigating residual covariance we introduce lower dimensional tensors, that successfully generalize to one side Kaluza-Klein gauge fields and to the other side extrinsic curvature and torsion of embedded...

Left juxtaposition of the atrial appendages is usually associated with cyanotic congenital heart disease. Recognition of this rare anomaly is important before therapeutic or surgical procedures that involve the atrial septum can be undertaken (for example, septostomy, the Mustard or Senning operation and the Fontan anastomosis). The diagnosis of left juxtaposition of the atrial appendages is mo...

Texture has long been regarded as spatial distributions of gray-level variation, and texture analysis has generally been connned to the 2-D image domain. Introducing the concept of \3-D world texture", this paper considers texture as a function of 3-D structures and proposes a set of \3-D textural features". The proposed 3-D features appear to have a great potential in terrain classiication. Ex...

Using drifting compound grating stimuli matched in energy and frequency spectrum, we previously showed that neurons in the primary visual cortex (V1) were tuned to line-like, edge-like, and intermediate one-dimensional features. Because these compound grating stimuli were drifting, allowing for potential interaction between shape and motion, we examine here the dependence of V1 feature tuning o...

Secondary structure of protein, such as α-helix and βsheet, is the general three-dimensional (3D) form of local segments. It can be identified from the 3D electron cryomicroscopy (cryo-EM) density images at medium resolutions (~5-10Å). A detected β-sheet can be represented by either the voxels of β-sheet density or by many piecewise polygons to compose a rough surface. However, none of these is...

The connectivity of so-called “special” and “general” grain boundaries at a quadruple node is known to be nonrandom as a result of crystallographic constraints. Although a quadruple node is a three-dimensional feature, there exist two-dimensional features which are topologically identical. Therefore, the distribution of these twodimensional features may be used to determine the three-dimensiona...

The integral image approach allows the optimal computation of Haar-based features for realtime recognition of objects in image sequences. This short note gives a direction to generalize the approach to high-dimensional images. We o er a formula for optimal computation of sums on high-dimensional rectangles. Integral Image Haar-Based Features High-Dimensional Image Möbius Inversion Formula

Twenty-eight patients with cardiac amyloidosis were studied by echocardiography -- 26 by M-mode and 13 by two-dimensional (2D) studies. All had heart failure and biopsy-proved amyloidosis, M-mode features included (1) normal left ventricular (LV) dimension in all; (2) thickened ventricular septum (88%), LV posterior wall (77%), and right ventricular (RV) anterior wall (79%); (3) decreased thick...

We propose a simple and straightforward way of creating powerful image representations via cross-dimensional weighting and aggregation of deep convolutional neural network layer outputs. We first present a generalized framework that encompasses a broad family of approaches and includes cross-dimensional pooling and weighting steps. We then propose specific non-parametric schemes for both spatia...