Visual speech-lip reading, making the computer understands what do speakers want to express through observing the lip direction of them. The most simply method of lip reading in early stage is to compare between chara...
详细信息
Visual speech-lip reading, making the computer understands what do speakers want to express through observing the lip direction of them. The most simply method of lip reading in early stage is to compare between characters from the frozen pictures and templates being stored. It neglects the character is changing with time. This method is very simply, but it only can classify the simple elements not the words, so it couldn't render great serves to speech recognition. Afterwards the adoption of behavioral characteristics is becoming more widespread. Because of the superior of Hidden Markov model (HMM), it can be applied in speech recognition widely. In recent years, it is also used in the research of lip-reading recognition. The classical HMM model makes two hypotheses: hidden expropriation hypothesis: the state at t+1 is only conditioned by the state at t, not the state before; the expropriation hypothesis from hidden state to visible state: the visible state at t only conditioned by the hidden state at t, not the state before. Such kind of hypothesis is not very reasonable in some practical application (such as lip-reading). In some kind condition, the state at t is not only conditioned by t-1, but also t-2. Therefore this thesis revises the assumed condition classical HMM to derive a new HMM model and algorithm, and applying it into lip-reading recognition to increase the discrimination.
Visual voice lip-reading, so the computer can understand what the speakers want to express direction by looking at their lips. Lip reading is the easiest way to compare the early characters and templates from the froz...
详细信息
Visual voice lip-reading, so the computer can understand what the speakers want to express direction by looking at their lips. Lip reading is the easiest way to compare the early characters and templates from the frozen image is stored. It ignores the very nature and time changes. This method is very simple, but it's just simple elements can be classified, then it may not show significant speech recognition services. Behavior was characterized by more and more common. Because of the hidden Markov model is superior (HMM), which can be widely used in speech recognition. In recent years, is also used to lip reading identification. Classical HMM model, so that the two assumptions: hidden assumptions collected: in t+1 the state can only be in this country is not in the state before t; from the hidden visible state hypothesis: only by regulating the t hide the visible state, rather than the previous state. This hypothesis is not very useful in some applications (such as lip reading) is reasonable. Under certain conditions, in the t state not only limits the t-1, but also t-2. Therefore, this study modified the assumptions of the classical HMM to derive a new HMM model and algorithms, and applied to the lip-reading recognition is increasing discrimination.
Aiming at the problem of multi-label classification, a multi-label classification algorithm based on label-specific features is proposed in this paper. In this algorithm, we compute feature density on the positive and...
详细信息
Aiming at the problem of multi-label classification, a multi-label classification algorithm based on label-specific features is proposed in this paper. In this algorithm, we compute feature density on the positive and negative instances set of each class firstly and then select mk features of high density from the positive and negative instances set of each class, respectively; the intersec- tion is taken as the label-specific features of the corresponding class. Finally, multi-label data are classified on the basis of la- bel-specific features. The algorithm can show the label-specific features of each class. Experiments show that our proposed method, the MLSF algorithm, performs significantly better than the other state-of-the-art multi-label learning approaches.
Max Restricted Path Consistency (maxRPC) is a promising domain filtering consistency for binary constraints. In existing algorithms, the process of searching for PC-witnesses requires most constraint checks. And the c...
详细信息
Topological relations between spatial objects have played important roles in spatial representing and reasoning. Various models for expressing topological relations have been put forward. The Region Connection Calculu...
详细信息
Among those researches in Deep Web, compared to research of data extraction which is more mature, the research of data annotation is still at its preliminary stage. Currently, although the approach of applying ontolog...
详细信息
作者：
LI YONGXU LUCollege of Mathematics
Key Laboratory of Symbolic computation and Knowledge Engineering of Ministry of EducationJilin UniversityChangchun130012
In this paper,we study the persistence of lower dimensional tori for random Hamiltonian systems,which shows that majority of the unperturbed tori persist as Cantor fragments of lower dimensional ones under small *** t...
详细信息
In this paper,we study the persistence of lower dimensional tori for random Hamiltonian systems,which shows that majority of the unperturbed tori persist as Cantor fragments of lower dimensional ones under small *** this result,we can describe the stability of the non-autonomous dynamic systems.
We focus on a special class of ideal projectors, subspaces, which possesses two classes of D -invariant polynomial subspaces. The first is a classical type, while the second is a new class. With matrix computation, we...
We focus on a special class of ideal projectors, subspaces, which possesses two classes of D -invariant polynomial subspaces. The first is a classical type, while the second is a new class. With matrix computation, we discretize this class of ideal projectors into a sequence of Lagrange projectors.
Most contemporary database systems query optimizers exploit System-R's bottom-up dynamic programming method (DP) to find the optimal query execution plan (QEP) without evaluating redundant subplans. The distinguis...
详细信息
Wireless communication, signal broadcasting, strictly limited resources in tags make RFID systems being confronted with many security attacks and privacy disclosure threats. In this paper, an efficient lightweight RFI...
详细信息
Wireless communication, signal broadcasting, strictly limited resources in tags make RFID systems being confronted with many security attacks and privacy disclosure threats. In this paper, an efficient lightweight RFID mutual authentication protocol is proposed, this protocol only requires O(1) work to identify and authenticate a tag in the backend server and is particularly suitable for the low-cost RFID systems because only one-way hash function, XOR operation and concatenation operation are needed in tags. The security and performance of this protocol are analyzed as well by comparing with the related RFID authentication protocols.
暂无评论