A deep descriptor for cross-tasking EEG-based recognition

A Contour Based Descriptor for Object Recognition

The paper deals with the extraction of features for the recognition of bidimensional objects. A set of invariants that is based on the contours of the objects is proposed. The invariance to rotation is founded on the properties of the Fourier Transform, applied to a circular signature of the contour. The specific method of creating the signature allows concave objects to be described. Interior ...

Deep Self-Convolutional Activations Descriptor for Dense Cross-Modal Correspondence

We present a novel descriptor, called deep self-convolutional activations (DeSCA), designed for establishing dense correspondences between images taken under different imaging modalities, such as different spectral ranges or lighting conditions. Motivated by descriptors based on local self-similarity (LSS), we formulate a novel descriptor by leveraging LSS in a deep architecture, leading to bet...

Deep Self-correlation Descriptor for Dense Cross-Modal Correspondence

A Segmentation Descriptor for Recognition

We present a segmentation-based image descriptor and similarity measure and apply them for object recognition. A segmentation gives a middle-level image representation which is robust to the intensity and structure variations that cause problems for matching based on low-level descriptors. A segmentation descriptor can give more accurate and efficient matching of boundaries than methods based o...

A novel Local feature descriptor using the Mercator projection for 3D object recognition

Point cloud processing is a rapidly growing research area of computer vision. Introducing of cheap range sensors has made a great interest in the point cloud processing and 3D object recognition. 3D object recognition methods can be divided into two categories: global and local feature-based methods. Global features describe the entire model shape whereas local features encode the neighborhood ...