The present paper describes the identification of prominent characters and their adjectives from Indian mythological epic, Mahabharata, written in English texts. However, in contrast to the traditional approaches of named entity identification, the present system extracts hidden attributes associated with each of the characters (e.g., character adjectives). We observed distinct phrase level lin...

Traditional CNN-based text recognition problems have either been single-character classification like MNIST or complex, stateful full-word classifications like the LSTM approach in Graves et al. I explore a stateless approach to word recognition by using Faster R-CNN’s region proposal network in conjunction with a heuristic to produce word labels from region proposals. While the results are pro...

We present an approach for animating characters using inverse kinematics (IK) handles that allows for intuitive keyframing of physically plausible motion. Specifically, we extend traditional IK and keyframing to include center-of-mass (CM) and inertia handles along with physically based templates to help an animator produce trajectories that respect physics during dynamic activities, such as sw...

Since the implementation of the "NCL System for Retrieval of Digitized Images of Ancient Books" much work has been spent on further development of its functionality, taking human interface design and ease of use as primary consideration. This article introduces new features that are available since the completion of this development stage in 2002, namely the handling of missing characters, conv...

There are a number of circumstances in which an escape mechanism is needed in conjunction with a protocol to encode characters that cannot be represented or transmitted directly. With ASCII coding, the traditional escape has been either the decimal or hexadecimal numeric value of the character, written in a variety of different ways. The move to Unicode, where characters occupy two or more octe...

We present the design and implementation of bounded-length sequence (BLS) variables for a CP solver. The domain of a BLS variable is represented as the combination of a set of candidate lengths and a sequence of sets of candidate characters. We show how this representation, together with requirements imposed by propagators, affects the implementation of BLS variables for a copying CP solver, mo...

The interactive approach of character animation requires sophisticated motion synthesis algorithms to generate poses and motions for unpredictable events. Traditional animation techniques used alone cannot produce the needed degree of realism or the degree of controllability required for virtual characters. Motion-capture techniques and the physical simulation of characters, when combined, offe...

In this paper, we propose new methods to learn Chinese word representations. Chinese characters are composed of graphical components, which carry rich semantics. It is common for a Chinese learner to comprehend the meaning of a word from these graphical components. As a result, we propose models that enhance word representations by character glyphs. The character glyph features are directly lea...

This paper defines the requirements for Autonomous Digital Actor’s cognitive process, in order to allow them to suggest plans of plausible actions in computer animations. An architecture is proposed, adapted from a traditional EBDI model considering those requirements. This kind of characters are inspired by real human actors and know how to act autonomously following a script, making the creat...

Signboards and billboards provide a challenge to image seg¬mentation methods, since these images may also have pictures and graphical objects, apart from text objects. Methods that often succeed in more traditional text block segmentation situations do not perform well here since estimation of text lines and character widths etc fail due to the short sample sizes. Further, extraction of charact...