Sketch recognition

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

Exploiting transfer learning for personalized view invariant gesture recognition

Abstract: A robust gesture recognition system is an essential component in many human-computer interaction applications. In particular, the widespread adoption of portable devices and the diffusion of autonomous systems with limited power and load capacity has increased the need of developing efficient recognition algorithms which operates on video streams recorded from low cost devices and which can cope with the challenging issue of point of view changes. A further challenge arises as different users tend to perform the same gesture with different styles and speeds. Thus a classifier trained with gestures data of certain set of users may work poorly when data from other users are being processed. However, as often a mobile device or a robot are intended to be used by a single or by a small group of people, it would be desirable to have a gesture recognition system designed specifically for these users. In this paper we introduce a novel approach to face the problems of view-invariance and user personalization in the context of gesture interaction systems. More specifically, we propose a domain adaptation framework based on a feature space augmentation approach operating on robust view-invariant Self Similarity Matrix descriptors. To prove the effectiveness of our method a dataset corresponding to 17 users performing 10 different gestures under 3 point of views is collected and an extensive experimental evaluation is performed.

Using Entropy to Identify Shape and Text in Hand Drawn Diagrams

Abstract: Most sketch recognition systems are accurate in recognizing either text or shape (graphic) ink strokes, but not both. Distinguishing between shape and text strokes is, therefore, a critical task in recognizing hand drawn digital ink diagrams which commonly contain many text labels and annotations. We have found the ‘entropy rate’ to be an accurate criterion of classification. We found that the entropy rate is significantly higher for text strokes compared to shape strokes and can serve as a distinguishing factor between the two. Using entropy values, our system produced a correct classification rate of 92.06% on test data belonging to diagrammatic domain for which the threshold was trained on. It also performed favorably on data for which no training examples at all were supplied.

Gesture Modelling and Recognition by Integrating Declarative Models and Pattern Recognition Algorithms

Abstract: Gesture recognition approaches based on computer vision and machine learning mainly focus on recognition accuracy and robustness. Research on user interface development focuses instead on the orthogonal problem of providing guidance for performing and discovering interactive gestures, through compositional approaches that provide information on gesture sub-parts. We make a first step toward combining the advantages of both approaches. We introduce DEICTIC, a compositional and declarative gesture description model which uses basic Hidden Markov Models (HMMs) to recognize meaningful pre-defined primitives (gesture sub-parts), and uses a composition of basic HMMs to recognize complex gestures. Preliminary empirical results show that DEICTIC exhibits a similar recognition performance as “monolithic” HMMs used in state-of-the-art vision-based approaches, retaining at the same time the advantages of declarative approaches.

The Human Visual Recognition Ability for Incomplete Letters

Abstract: The human vision system has high-level recognition abilities and can efficiently and accurately identify an object based on object parts. In the present study, a letter recognition experiment was designed to evaluate recognition of incomplete objects by humans. Five subjects were asked to identify partially erased English letters displayed for short time periods. We found that features and correlations between features significantly influence letter identification. An algorithm based on information entropy is proposed to quantitatively evaluate the importance of letter features and the similarities between letters.

A novel human hand finger gesture recognition using machine learning

Abstract: Human-Computer Interaction (HCI) using intelligent artificial computing interface is a fast emerging and revolutionary field of study of computer vision. This present study is concerned with making computers responsive to human gestures and postures. In this paper a simple alternative method for hand gesture recognition system has been proposed. The system takes various fingers postures and try to recognize them using machine learning. A pattern of gestures is trained and tested to show the results using linear artificial neural network.