SPEECH RECOGNITION IN VODIS II.

VODIS II, a research system in which recognition is based on the conventional one-pass connected-word algorithm extended in two ways, is described. Syntactic constraints can now be applied directly via context-free-grammar rules, and the algorithm generates a lattice of candidate word matches rather than a single globally optimal sequence. This lattice is then processed by a chart parser and an intelligent dialogue controller to obtain the most plausible interpretations of the input. A key feature of the VODIS II architecture is that the concept of an abstract word model allows the system to be used with different pattern-matching technologies and hardware. The current system implements the word models on a real-time dynamic-time-warping recognizer.

Applications of stochastic context-free grammars using the Inside-Outside algorithm

This paper describes two applications in speech recognition of the use of stochastic context-free grammars (SCFGs) trained automatically via the Inside-Outside Algorithm. First, SCFGs are used to model VQ encoded speech for isolated word recognition and are compared directly to HMMs used for the same task. It is shown that SCFGs can model this low-level VQ data accurately and that a regular grammar based pre-training algorithm is effective both for reducing training time and obtaining robust solutions. Second, an SCFG is inferred from a transcription of the speech used to train a phoneme-based recognizer in an attempt to model phonotactic constraints. When used as a language model, this SCFG gives improved performance over a comparable regular grammar or bigram. © 1991.

Exploiting variable-width features in large vocabulary speech recognition

The use of variable-width features (prosodics, broad structural information etc.) in large vocabulary speech recognition systems is discussed. Although the value of this sort of information has been recognized in the past, previous approaches have not been widely used in speech systems because either they have not been robust enough for realistic, large vocabulary tasks or they have been limited to certain recognizer architectures. A framework for the use of variable-width features is presented which employs the N-Best algorithm with the features being applied in a post-processing phase. The framework is flexible and widely applicable, giving greater scope for exploitation of the features than previous approaches. Large vocabulary speech recognition experiments using TIMIT show that the application of variable-width features has potential benefits.

The modified Kanerva model for automatic speech recognition

A parallel processing network derived from Kanerva's associative memory theory Kanerva 1984 is shown to be able to train rapidly on connected speech data and recognize further speech data with a label error rate of 0·68%. This modified Kanerva model can be trained substantially faster than other networks with comparable pattern discrimination properties. Kanerva presented his theory of a self-propagating search in 1984, and showed theoretically that large-scale versions of his model would have powerful pattern matching properties. This paper describes how the design for the modified Kanerva model is derived from Kanerva's original theory. Several designs are tested to discover which form may be implemented fastest while still maintaining versatile recognition performance. A method is developed to deal with the time varying nature of the speech signal by recognizing static patterns together with a fixed quantity of contextual information. In order to recognize speech features in different contexts it is necessary for a network to be able to model disjoint pattern classes. This type of modelling cannot be performed by a single layer of links. Network research was once held back by the inability of single-layer networks to solve this sort of problem, and the lack of a training algorithm for multi-layer networks. Rumelhart, Hinton & Williams 1985 provided one solution by demonstrating the "back propagation" training algorithm for multi-layer networks. A second alternative is used in the modified Kanerva model. A non-linear fixed transformation maps the pattern space into a space of higher dimensionality in which the speech features are linearly separable. A single-layer network may then be used to perform the recognition. The advantage of this solution over the other using multi-layer networks lies in the greater power and speed of the single-layer network training algorithm. © 1989.

Face recognition from video

In spite of over two decades of intense research, illumination and pose invariance remain prohibitively challenging aspects of face recognition for most practical applications. The objective of this work is to recognize faces using video sequences both for training and recognition input, in a realistic, unconstrained setup in which lighting, pose and user motion pattern have a wide variability and face images are of low resolution. In particular there are three areas of novelty: (i) we show how a photometric model of image formation can be combined with a statistical model of generic face appearance variation, learnt offline, to generalize in the presence of extreme illumination changes; (ii) we use the smoothness of geodesically local appearance manifold structure and a robust same-identity likelihood to achieve invariance to unseen head poses; and (iii) we introduce an accurate video sequence "reillumination" algorithm to achieve robustness to face motion patterns in video. We describe a fully automatic recognition system based on the proposed method and an extensive evaluation on 171 individuals and over 1300 video sequences with extreme illumination, pose and head motion variation. On this challenging data set our system consistently demonstrated a nearly perfect recognition rate (over 99.7%), significantly outperforming state-of-the-art commercial software and methods from the literature. © Springer-Verlag Berlin Heidelberg 2006.

Segmentation and recognition using structure from motion point clouds

We propose an algorithm for semantic segmentation based on 3D point clouds derived from ego-motion. We motivate five simple cues designed to model specific patterns of motion and 3D world structure that vary with object category. We introduce features that project the 3D cues back to the 2D image plane while modeling spatial layout and context. A randomized decision forest combines many such features to achieve a coherent 2D segmentation and recognize the object categories present. Our main contribution is to show how semantic segmentation is possible based solely on motion-derived 3D world structure. Our method works well on sparse, noisy point clouds, and unlike existing approaches, does not need appearance-based descriptors. Experiments were performed on a challenging new video database containing sequences filmed from a moving car in daylight and at dusk. The results confirm that indeed, accurate segmentation and recognition are possible using only motion and 3D world structure. Further, we show that the motion-derived information complements an existing state-of-the-art appearance-based method, improving both qualitative and quantitative performance. © 2008 Springer Berlin Heidelberg.

Silhouette-based object phenotype recognition using 3D shape priors

This paper tackles the novel challenging problem of 3D object phenotype recognition from a single 2D silhouette. To bridge the large pose (articulation or deformation) and camera viewpoint changes between the gallery images and query image, we propose a novel probabilistic inference algorithm based on 3D shape priors. Our approach combines both generative and discriminative learning. We use latent probabilistic generative models to capture 3D shape and pose variations from a set of 3D mesh models. Based on these 3D shape priors, we generate a large number of projections for different phenotype classes, poses, and camera viewpoints, and implement Random Forests to efficiently solve the shape and pose inference problems. By model selection in terms of the silhouette coherency between the query and the projections of 3D shapes synthesized using the galleries, we achieve the phenotype recognition result as well as a fast approximate 3D reconstruction of the query. To verify the efficacy of the proposed approach, we present new datasets which contain over 500 images of various human and shark phenotypes and motions. The experimental results clearly show the benefits of using the 3D priors in the proposed method over previous 2D-based methods. © 2011 IEEE.

Achieving robust face recognition from video by combining a weak photometric model and a learnt generic face invariant

In spite of over two decades of intense research, illumination and pose invariance remain prohibitively challenging aspects of face recognition for most practical applications. The objective of this work is to recognize faces using video sequences both for training and recognition input, in a realistic, unconstrained setup in which lighting, pose and user motion pattern have a wide variability and face images are of low resolution. The central contribution is an illumination invariant, which we show to be suitable for recognition from video of loosely constrained head motion. In particular there are three contributions: (i) we show how a photometric model of image formation can be combined with a statistical model of generic face appearance variation to exploit the proposed invariant and generalize in the presence of extreme illumination changes; (ii) we introduce a video sequence re-illumination algorithm to achieve fine alignment of two video sequences; and (iii) we use the smoothness of geodesically local appearance manifold structure and a robust same-identity likelihood to achieve robustness to unseen head poses. We describe a fully automatic recognition system based on the proposed method and an extensive evaluation on 323 individuals and 1474 video sequences with extreme illumination, pose and head motion variation. Our system consistently achieved a nearly perfect recognition rate (over 99.7% on all four databases). © 2012 Elsevier Ltd All rights reserved.

User target intention recognition from cursor position using kalman filter

This paper discusses user target intention recognition algorithms for pointing - clicking tasks to reduce users' pointing time and difficulty. Predicting targets by comparing the bearing angles to targets proposed as one of the first algorithms [1] is compared with a Kalman Filter prediction algorithm. Accuracy and sensitivity of prediction are used as performance criteria. The outcomes of a standard point and click experiment are used for performance comparison, collected from both able-bodied and impaired users. © 2013 Springer-Verlag Berlin Heidelberg.

The multi-weight neuron with geometry algorithm and its application

A novel approach for multi-dimension signals processing, that is multi-weight neural network based on high dimensional geometry theory, is proposed. With this theory, the geometry algorithm for building the multi-weight neuron is mentioned. To illustrate the advantage of the novel approach, a Chinese speech emotion recognition experiment has been done. From this experiment, the human emotions are classified into 6 archetypal classes: fear, anger, happiness, sadness, surprise and disgust. And the amplitude, pitch frequency and formant are used as the feature parameters for speech emotion recognition. Compared with traditional GSVM model, the new method has its superiority. It is noted that this method has significant values for researches and applications henceforth.

The application of DBF neural networks for object recognition

On the basis of DBF nets proposed by Wang Shoujue, the model and properties of DBF neural network were discussed in this paper. When applied in pattern recognition, the algorithm and implement on hardware were presented respectively. We did experiments on recognition of omnidirectionally oriented rigid objects on the same level, using direction basis function neural networks, which acts by the method of covering the high dimensional geometrical distribution of the sample set in the feature space. Many animal and vehicle models (even with rather similar shapes) were recognized omnidirectionally thousands of times. For total 8800 tests, the correct recognition rate is 98.75%, the error rate and the rejection rate are 0.5% and 1.25% respectively. (C) 2003 Elsevier Inc. All rights reserved.

Geometrical learning, descriptive geometry, and biomimetic pattern recognition

Studies on learning problems from geometry perspective have attracted an ever increasing attention in machine learning, leaded by achievements on information geometry. This paper proposes a different geometrical learning from the perspective of high-dimensional descriptive geometry. Geometrical properties of high-dimensional structures underlying a set of samples are learned via successive projections from the higher dimension to the lower dimension until two-dimensional Euclidean plane, under guidance of the established properties and theorems in high-dimensional descriptive geometry. Specifically, we introduce a hyper sausage like geometry shape for learning samples and provides a geometrical learning algorithm for specifying the hyper sausage shapes, which is then applied to biomimetic pattern recognition. Experimental results are presented to show that the proposed approach outperforms three types of support vector machines with either a three degree polynomial kernel or a radial basis function kernel, especially in the cases of high-dimensional samples of a finite size. (c) 2005 Elsevier B.V. All rights reserved.

Manifold covering theory in biomimetic pattern recognition

We describe a new model which is based on the concept of cognizing theory. The method identifies subsets of the data which are embedded in arbitrary oriented lower dimensional space. We definite manifold covering in biomimetic pattern recognition, and study its property. Furthermore, we propose this manifold covering algorithm based on Biomimetic Pattern Recognition. At last, the experimental results for face recognition demonstrates that the correct rejection rate of the test samples excluded in the classes of training samples is very high and effective.

Information geometry theory of high-dimension space and application for speaker independent continuous digit speech recognition

In the light of descriptive geometry and notions in set theory, this paper re-defines the basic elements in space such as curve and surface and so on, presents some fundamental notions with respect to the point cover based on the High-dimension space (HDS) point covering theory, finally takes points from mapping part of speech signals to HDS, so as to analyze distribution information of these speech points in HDS, and various geometric covering objects for speech points and their relationship. Besides, this paper also proposes a new algorithm for speaker independent continuous digit speech recognition based on the HDS point dynamic searching theory without end-points detection and segmentation. First from the different digit syllables in real continuous digit speech, we establish the covering area in feature space for continuous speech. During recognition, we make use of the point covering dynamic searching theory in HDS to do recognition, and then get the satisfying recognized results. At last, compared to HMM (Hidden Markov models)-based method, from the development trend of the comparing results, as sample amount increasing, the difference of recognition rate between two methods will decrease slowly, while sample amount approaching to be very large, two recognition rates all close to 100% little by little. As seen from the results, the recognition rate of HDS point covering method is higher than that of in HMM (Hidden Markov models) based method, because, the point covering describes the morphological distribution for speech in HDS, whereas HMM-based method is only a probability distribution, whose accuracy is certainly inferior to point covering.

Research on continuous speech recognition with small vocabulary based on biomimetic pattern recognition

Based on biomimetic pattern recognition theory, we proposed a novel speaker-independent continuous speech keyword-spotting algorithm. Without endpoint detection and division, we can get the minimum distance curve between continuous speech samples and every keyword-training net through the dynamic searching to the feature-extracted continuous speech. Then we can count the number of the keywords by investigating the vale-value and the numbers of the vales in the curve. Experiments of small vocabulary continuous speech with various speaking rate have got good recognition results and proved the validity of the algorithm.