Sketch recognition

About: Sketch recognition is a research topic. Over the lifetime, 1611 publications have been published within this topic receiving 40284 citations.

Arithmetic of Gesture Recognition Using Multi-Scale Model and Moment-Describe

Abstract: This paper indicates an approach of gesture recognition with single camera We get the basic parameters of recognition through initialization of users, and get the region of gesture of user through video tracking, and then in this region we get characters using moment-describe and multi-scale model At last the gesture recognition using classified rule is realized The experimental results show that our arithmetic is fit for different users, and the recognition is exact

User gesture recognition

Abstract: A method and device for gesture recognition, wherein the gesture is executed by a user in a gesture region which may be defined relative to a display surface. In an embodiment, the gesture comprises a select gesture and the device comprises at least three cameras operating in the visual range where a first camera is used to determine a horizontal location of the select gesture and the other cameras are used to determine a vertical location thereof. A device for providing input to a computing device comprises a rectangular display having a viewing surface and at least three cameras having respective fields of view. A first camera and a second camera are located at respective adjacent corners of the display and a third camera is located at an edge of the display between the first and second cameras.

DSD: Depth Structural Descriptor for Edge-Based Assistive Navigation

Abstract: Structural edge detection is the task of finding edges between significant surfaces in a scene. This can underpin many computer vision tasks such as sketch recognition and 3D scene understanding, and is important for conveying scene structure for navigation with assistive vision. Identifying structural edges from a depth image can be challenging because surface structure that differentiates edges is not well represented in this format. We derive a depth input encoding, the Depth Surface Descriptor (DSD), that captures the first order properties of surfaces, allowing for improved classification of surface geometry that corresponds to structural edges. We apply the DSD feature to salient edge detection on RGB-D images using a fully convolutional neural network with deep supervision. We evaluate our method on both a new RGB-D dataset containing prosthetic vision scenarios, and the SUNRGBD dataset, and show that our approach produces improved performance compared to existing methods by 4%.

A handwritten recognition for free style Marathi script using genetic algorithm

Abstract: Automatic character recognition of handwritten numerals and characters has been an active subject of research due to its importance on industrial as well as educational platform. The off-line handwritten character recognition is an active area for research towards the new techniques that would help to improve recognition accuracy. Now a day's looking forward for rapidly growing technologies, with its applications in various fields of automation industry pattern recognition in which specially character recognition forms core research area within engineering and computer science disciplines. This paper investigates the method for free style Marathi off-line handwritten Character Recognition with the genetic algorithm approach and also discuss some existing methods for it.