EP2828797A1 - Dispositif et procede d'identification d'un mouvement cyclique, programme d'ordinateur correspondant - Google Patents
Dispositif et procede d'identification d'un mouvement cyclique, programme d'ordinateur correspondantInfo
- Publication number
- EP2828797A1 EP2828797A1 EP13715344.1A EP13715344A EP2828797A1 EP 2828797 A1 EP2828797 A1 EP 2828797A1 EP 13715344 A EP13715344 A EP 13715344A EP 2828797 A1 EP2828797 A1 EP 2828797A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- cyclic
- hmm
- type
- sequence
- statistical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
Definitions
- the present invention relates to a device for identifying a type of cyclic movement among a set of types of possible cyclic movements of a physical system observed by at least one motion sensor. It also relates to a corresponding method and computer program.
- physical system any system producing a physical output observable by a sensor, the system being assumed a priori to be able to reproduce a predetermined number of cyclical movements modeled by the identification device.
- the observed physical system may for example be an animated object capable of being driven according to a cyclical movement that it is desired to identify, or even monitor to detect possible anomalies, using one or more sensors.
- the detectable cyclic movements are variable and the applications multiple.
- the invention relates to a device for identifying a type of cyclic movement among a set of possible types of cyclic movements of a physical system observed by at least one motion sensor, comprising:
- At least one motion sensor for providing a sequence of observation data of the physical system
- a computer connected to the sensor and to the storage means, programmed to analyze the sequence of observation data and to select one of the possible types of cyclical movements on the basis of the stored statistical model.
- the cyclic movement observed is a swimming activity.
- the motion sensor is at least one measurement axis and worn by a user.
- the calculator is more specifically programmed to analyze a sequence of observation data and to identify the type of swimming practiced by the user over time by using a unique Markov statistical model with N hidden states corresponding respectively to N types of strokes. identifiable.
- the identification of the type of swimming is possible but rather summary.
- different variants can be practiced by the same person or by different people.
- the model applied in this document does not allow to describe more precisely the type of swimming observed so that such variants can not be detected a priori, except to multiply the number of hidden states in the statistical model used by modeling a priori these different variants and thus also increase the risk of detection errors.
- a device for identifying a cyclic movement among a set of possible cyclical movements of a physical system observed by at least one motion sensor comprising:
- At least one motion sensor for providing a sequence of observation data of the physical system
- a computer connected to the sensor and to the storage means, programmed to analyze the sequence of observation data and select one of the possible cyclical movements on the basis of the stored statistical model
- the storage means comprise at least one statistical Markov model with hidden states by possible cyclic movement
- the computer is programmed to analyze a compatibility of the observation data sequence with each cyclic movement possible based on the statistical model that is associated with this cyclical movement,
- the computer is further programmed to provide a sequence of hidden states of the statistical model associated with the cyclic movement selected from the observation data sequence and the statistical model associated with the selected cyclic movement.
- a device instead of having a single Markov model with N hidden states corresponding to the N possible cyclic movements as would be taught in patent application FR 2 943 554, a device according to the invention provides for at least N state Markov models. hidden to identify the N cyclic movements. The modeling is therefore finer because then the hidden states of the same Markov model can model several components of the same cyclic movement.
- a sequence of hidden states of the statistical model associated with the cyclic motion selected from the sequence of observation data from the sensor a detailed temporal analysis of the identified motion and its components is available.
- each hidden-state Markov statistical model relating to a possible cyclic motion is a cyclic model represented by an oriented graph in which an oriented cycle of hidden states is imposed.
- each hidden state of each statistical model of the storage means corresponds to a predetermined sub-movement of the cyclic movement with which this statistical model is associated.
- a hidden state corresponding to a starting sub-movement of the cyclic motion with which this statistical model is associated is represented by a node of the corresponding oriented graph situated upstream of said oriented cycle
- a hidden state corresponding to a stop sub-movement of the cyclic movement with which this statistical model is associated is represented by a node of the corresponding oriented graph located downstream from said oriented cycle.
- a method for identifying a cyclic movement among a set of possible cyclic motions of a physical system observed by at least one sensor comprising:
- At least one statistical Markov model with hidden states by possible cyclic motion is stored in memory
- the selection is made by analyzing a compatibility of the sequence of observation data with each possible cyclical movement on the basis of the statistical model which is associated with this cyclical movement, the identification method further comprising the provision of a sequence of hidden states of the statistical model associated with the selected cyclic motion from the observation data sequence and the statistical model associated with the selected cyclic motion.
- the selection includes a comparison of probabilities for possible cyclical movements knowing the sequence of observation data, the probabilities being estimated based on the stored statistical models.
- the provision of said sequence of hidden states is performed by applying the Viterbi algorithm.
- the identification method as defined above further comprises the computation of statistical parameters or counting of said sequence of hidden states provided.
- each hidden-state Markov statistical model relating to a possible cyclic movement being a cyclic model in which an oriented cycle of hidden states is imposed
- the calculated statistics or counting parameters comprise at least one of the elements. of the set consisting of a number of state cycles hidden in said provided hidden state sequence, durations of each cycle, an average of the durations of each cycle, a variance of the durations of each cycle and a pause period in a cycle.
- FIG. 1 diagrammatically represents the general structure of an identification device according to one embodiment of the invention
- FIG. 2 illustrates an example of a hidden state statistical model of Markov having constraints of transitions between hidden states forming a cycle and which can be taken into account by the identification device of FIG. 1,
- FIG. 3 illustrates a particular use of the identification device of FIG. 1,
- FIG. 4 illustrates the successive steps of an identification method implemented by the device of FIG. 1, and
- FIG. 5 illustrates, by means of a diagram, a result of the identification method of FIG. 4 obtained on the basis of a cyclic statistical model such as that of FIG. 2.
- the device 10 shown in FIG. 1 is a device for identifying a cyclic movement among a set of possible cyclical movements of a physical system observed by at least one motion sensor. It comprises for this purpose an observation module 12, a processing module 14 and an interface module 16.
- the observation module 12 comprises one or more sensors represented by the unique reference 18 for the observation of the physical system.
- the sensor 18 may for example comprise a motion sensor with one, two or three measurement axes, in particular a 3D accelerometer carried by a person, for the identification of a cyclic movement of this person, for example in the context of a walking, running or swimming activity of that person,
- it may comprise a motion sensor for determining the activity of a mobile system in a set of possible repetitive or cyclic activities
- the motion sensor may comprise an accelerometer, a gyrometer or a magnetometer, - etc.
- the sensor 18 may also comprise several sensors each providing observations which, combined, make it possible to envisage detecting more complex cyclic movements.
- observation data can be directly derived from a sampling of the observation signal or obtained after one or more treatments, in particular one or more filtering, of this signal. It is thus understood that the observation data can be mono or multivalued, even when only one sensor 18 is available.
- the processing module 14 is an electronic circuit, for example that of a computer. It comprises storage means 20, for example a memory type RAM, ROM or other, in which are stored the parameters of several statistical models Markov hidden state.
- Each cyclic motion S-1, S-N expected to be detectable by the detection device 10 using the sensor 18 is modeled by a corresponding concealed state Markov statistical model denoted HMM-1, HMM-N.
- HMM-1 concealed state Markov statistical model
- a first HMM-1 model can be breaststroke
- a second HMM-2 model can match the crawl
- a third HMM-3 model can match the backstroke, etc.
- HMM-n Any of the stored hidden state statistical Markov models, denoted HMM-n and modeling the cyclic motion S-n, is defined by the following parameters:
- the probability law of each hidden state i of the model HMM-n can be chosen from the family normal laws. In this case, it is defined by its expectation ⁇ ⁇ and its variance ⁇ n
- ⁇ ⁇ is a vector comprising as many components
- ⁇ n ⁇ a matrix comprising as many rows and columns only values provided at each moment.
- HMM-n statistical model some constraints may be imposed in the HMM-n statistical model, including constraints on transitions,] a hidden state to another, some may be prohibited.
- the HMM-n model can be itself of cyclic type.
- FIG. 2 An example cyclic HMM-n cyclic model with three cyclic hidden states K2, K3, K4 and five concealed states in total K1 to K5 is illustrated in FIG. 2. It is suitable for modeling a physical activity of a person, detectable at using an accelerometer sensor, most of these activities being periodic or pseudoperiodic. For example, walking or swimming involves successions of sub-movements, respectively limbs and head, which are repeated.
- the first state K1 represents the commitment of the periodic or pseudoperiodic activity considered (for example a thrust on the edge of a basin in the case of a swimming)
- the three successive states K2, K3 and K4 represent the periodic phase or pseudoperiodic of the activity itself, and the state K5, the end or exit of the activity.
- a 1 t1 1 - ⁇ ⁇ 5
- a ⁇ 2 ⁇ 1
- a 2.2 1 - ⁇ 2 - ⁇ 2
- a 2.3 ⁇ 2
- a 2.5 ⁇ 2
- a 3.3 1 - ⁇ 3 - ⁇ 3
- a 3 4 ⁇ 3
- a 3 5 ⁇ 3
- a 4 1 4 1 - ⁇ 4 - ⁇ 4
- a 4 2 ⁇ 4
- a cyclic HMM-n model with Cn hidden states of which Cn-2 are cyclic is constrained as follows:
- a breaststroke swim could be represented by a cyclic pattern with five cyclic hidden states (K2, K3, K4, K5, K6), an engagement state (K1) upstream of the cycle and a state (K7) of output downstream of the cycle. would correspond to one of the following five successive sub-movements of the breaststroke:
- the arms spread widely, palms facing outward, finger tight, to form an angle of 45 ° to the axis head / feet; towards the end of the movement, the palms are oriented towards the bottom, to facilitate the exit of the head and the shoulders of the water,
- the memory 20 may further store, in association with each HMM-n model, one or more L-n training sequences.
- Each training sequence of the HMM-n model is in fact an observation sequence provided by the sensor 18, but it is known that it was extracted from the observation of the physical system while it reproduced the cyclic motion Sn.
- the formation of a base of training sequences can for example be obtained by performing an observation method such as that described in the patent application FR 2 943 554. It can also be obtained empirically.
- a learning sequence Ln can be processed upon reception by the processing module 14, or stored in memory 20 in connection with the HMM-n model for further processing, for a configuration or reconfiguration of the identification device. 10 by updating parameters of the HMM-n model, as is for example detailed in the French patent application published under the number FR 2 964 223.
- the processing module 14 further comprises a computer 22, for example a central computer unit equipped with a microprocessor 24 and a storage space of at least one computer program 26. This computer 22, and more particularly the microprocessor 24, is connected to the sensor 18 and the memory 20.
- the computer program 26 performs three main functions illustrated by modules 28, 30 and 32 in FIG.
- the first function fulfilled by the identification module 28, for example in the form of at least one instruction loop, is a function of identification of a cyclic movement reproduced by the physical system, on receipt of a observation sequence O (0: M-1) provided by the sensor 18. This function will be detailed with reference to FIG. 4.
- the identification module 28 comprises a first module 28A for identifying a type of cyclic movement among several possible types.
- This first module 28A is programmed to select one of the possible cyclic movements S-1, SN by analyzing a compatibility of the observation sequence 0 (0: M-1) with each possible cyclic movement on the basis of of the statistical model that is associated with this cyclical movement.
- This compatibility analysis comprises, for example, a comparison of probabilities relating to these possible cyclical movements knowing the observation sequence, the probabilities being estimated on the basis of the statistical HMM-1 stored models,
- HMM-N HMM-N.
- the resolution of this selection using hidden-state Markov statistical models is well known and is one of the three major classes of problems solved by hidden Markov models, as mentioned in L. Rabiner's article entitled “A tutorial on Hidden Markov Models and selected applications in speech recognition," Proceedings of the IEEE, vol. 77, no. 2, pp. 257-286, February 1989.
- This is the first class of problems mentioned in this article on page 261 and whose resolution is detailed on pages 262 and 263.
- the identification module 28 includes a second module 28B for temporal analysis of the cyclic movement selected by the first module 28A.
- This second module 28B is programmed to provide a sequence of hidden states X (0: M-1) of the statistical model associated with the cyclic motion selected from the observation sequence and the statistical model associated with the selected cyclic movement.
- the resolution of this provision of a hidden state sequence X (0: M-1) using a hidden-state Markov statistical model is well known and is also one of three major classes of problems solved by the hidden Markov models, as mentioned in the article by L. Rabiner mentioned above. This is the second class of problems mentioned in this article on page 261 and whose resolution is detailed on pages 263 and 264.
- this second class of problems it is desired to provide the sequence of hidden states X (0: M -1) corresponding best, according to a certain optimality criterion, to the observation sequence X (0: M-1) knowing the Markov statistical model selected for this observation sequence.
- Several solutions are known for solving this second class of problems. One of them is to apply the Viterbi algorithm as described in the L. Rabiner article mentioned earlier. From the hidden state sequence provided in this manner, also called the Viterbi path, the second time analysis module 28B is further programmed to compute statistical or counting parameters. For example, given the cyclic nature of the HMM-1, HMM-N HMM statistic models, these statistics or counting parameters may include a number of hidden state cycles in the hidden state sequence provided.
- the study of these statistical or counting parameters from the Viterbi path then makes it possible to access a more precise identification of the cyclical movement identified.
- the second function fulfilled by the recording module 30, for example in the form of an instruction loop, is a function of recording, in the memory 20, an observation sequence in relation to the one of the possible cyclic movements S-1, SN. This observation sequence then becomes a learning sequence to be used for configuring or reconfiguring the identification device 10.
- the third function fulfilled by the configuration module 32, for example in the form of an instruction loop, is a configuration function of the identification device 10 by updating the parameters of at least one statistical model HMM stored in memory 20 by means of a training sequence or a set of corresponding learning sequences Ln.
- This function is for example described in the patent application FR 2 964 223 and will not be detailed. It can, however, advantageously be adapted according to the specific constraints imposed by the cyclic models.
- the interface module 16 may comprise a mode selector 34 controlled by a user, in particular the person carrying the identification device 10 itself, when the observed physical system is a person.
- the identification device 10 operates by default in identification mode, thus executing the identification module 28.
- the identification device 10 may momentarily enter record mode, when an observation sequence associated with a known cyclical movement reproduced by the observed physical system is provided by the sensor 18 and must be recorded as a training sequence in the memory 20.
- the identification device may then comprise a recording interface 36, with the aid of which the user defines the observation sequence (for example marking its beginning and end) and associates it with one of the possible cyclical movements.
- the recording interface 36 may comprise, in a conventional manner, a screen and / or input means.
- the identification device 10 can momentarily go into configuration mode, when the user feels that he has enough training sequences in memory 20 to improve the adaptation of the detection device 10 to the observed physical system.
- observation modules 12, processing 14 and interface 16 are structurally separable.
- the identification device 10 may be designed in one piece or in several separate hardware elements interconnected by means of data transmission with or without wire.
- processing modules 14 and possibly the interface modules 16 can be implemented by computer. Only the observation module 12 is necessarily in the vicinity or in contact with the physical system observed since it comprises the sensor or sensors.
- FIG. 3 a particularly compact embodiment is illustrated for an application for identifying a cyclic movement of a person 40.
- the identification device 10 is entirely integrated in a housing. 42 carried by the person.
- the sensor is for example a 3D accelerometer and cyclic movements observed are for example types of swimming.
- the housing 42 is for example firmly held on an arm of the person 40 by means of a bracelet 44, so that the detection device 10 is worn as a watch.
- the housing 42 could be held on the front of the person 40 by means of a strip.
- the housing 42 could be held on one of the legs of the person 40.
- FIG. 4 illustrates the successive steps of a method of configuring the device 10 and of identifying observation sequences using the device 10 once configured.
- a base of training sequences is constituted. It can be constituted by execution of the recording module 30 and / or by execution of an observation method such as that described in the patent application FR 2 943 554.
- a training sequence is in fact an observation sequence in relation to one of the possible cyclic movements S-1, SN.
- the data of an observation sequence can be directly derived from a sampling of the observation signal provided by the sensor 18 or obtained after one or more treatments, in particular one or more filtering, of this signal. If additional treatment is deemed necessary on any observation sequence before confronting it with the HMM-1, HMM-N cyclic statistical models, in particular to extract a sequence of particular characteristics, then this additional processing can be performed by the computer 22 during an optional step 102 .
- a step 104 the device 10 is configured or reconfigured using the base of the training sequences.
- the execution of this step will not be detailed. As indicated above, this can be done by an adaptation of the teaching disclosed in the patent application FR 2 964 223. Alternatively, this step can also be performed empirically and not reconfigurable.
- the cyclical statistical models HMM-1, HMM-N are thus up to date and exploitable.
- the steps 100 to 104 detailed above allow the configuration of the identification device 10 so as to make it operational and can be repeated as often as desired.
- the actual identification method implemented by the device 10, and even more precisely by its identification module 28, comprises a step 200 of reception, by the computer 22 of the processing module 14, of a sequence of observation of the physical system provided by the sensor 18.
- step 202 If additional processing is deemed necessary on the received observation sequence before comparing it with the cyclic statistical models HMM-1, HMM-N, in particular to extract a sequence of particular characteristics, then this additional processing can be performed by the calculator 22 in an optional step 202 identical to step 102
- the compatibility of the observation sequence with each cyclic statistical model HMM-n is estimated, as indicated previously in the description of the first module 28A.
- the HMM-i cyclic statistical model maximizing the calculated probabilities is selected and the corresponding cyclical movement is identified.
- the sequence of hidden states representing at best the sequence of observation is calculated and supplied at the output of the computer 22, as indicated previously in the description of the second module 28B, that is to say by application of the Viterbi algorithm as described in the article by L. Rabiner mentioned above.
- the sequence of hidden states obtained reproduces these constraints and thus finely characterizes the sequence of the sub-movements of the identified cyclic motion. It can then be displayed on a screen, for example a screen of the identification device 10 if it has one, and be statistically analyzed. In particular, statistical or counting parameters as mentioned above can be evaluated and possibly displayed.
- FIG. 5 illustrates the assignment of hidden states to the samples of an observation sequence of a cyclic activity such as swimming, based on a cyclic pattern selected by the identification device 10 (step 204 ) to a state of engagement (state 1) and three cyclic states (states 2, 3 and 4) such as that of figure 2.
- the sampled observation sequence is illustrated by the curve C1
- the Viterbi path corresponding, provided by the identification device 10 (step 206) is illustrated by the curve C2.
- the statistical analysis of the Viterbi C2 path allows to analyze its irregularities of cycles, their number, their speed, etc.
- an identification device such as that described above allows identification and fine analysis of the cyclic movements reproduced by an observed physical system.
- the fact of providing the Viterbi path associated with an observation sequence whose cyclic motion it has reproduced has been identified makes it possible to access the temporal and statistical analysis of the subunits simply and with sufficient precision. movements of this identified movement.
- the detection device can be designed in very different forms since its observation modules 12, processing 14 and interface 16 are separable. Its design can thus be adapted to the intended application and the observed physical system.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Analysis (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FR1252493A FR2988501B1 (fr) | 2012-03-20 | 2012-03-20 | Dispositif et procede d'identification d'un mouvement cyclique, programme d'ordinateur correspondant |
| PCT/FR2013/050568 WO2013140075A1 (fr) | 2012-03-20 | 2013-03-18 | Dispositif et procede d'identification d'un mouvement cyclique, programme d'ordinateur correspondant |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP2828797A1 true EP2828797A1 (fr) | 2015-01-28 |
Family
ID=48083516
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP13715344.1A Withdrawn EP2828797A1 (fr) | 2012-03-20 | 2013-03-18 | Dispositif et procede d'identification d'un mouvement cyclique, programme d'ordinateur correspondant |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US9952043B2 (fr) |
| EP (1) | EP2828797A1 (fr) |
| FR (1) | FR2988501B1 (fr) |
| WO (1) | WO2013140075A1 (fr) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10973517B2 (en) | 2015-11-13 | 2021-04-13 | Intuitive Surgical Operations, Inc. | Stapler with composite cardan and screw drive |
| WO2020131290A1 (fr) | 2018-12-21 | 2020-06-25 | Intuitive Surgical Operations, Inc. | Ensembles articulation pour instruments chirurgicaux |
| WO2020131685A1 (fr) | 2018-12-21 | 2020-06-25 | Intuitive Surgical Operations, Inc. | Instruments chirurgicaux dotés de commutateurs pour désactiver et/ou identifier des cartouches d'agrafeuse |
| WO2020214258A1 (fr) | 2019-04-15 | 2020-10-22 | Intuitive Surgical Operations, Inc. | Cartouche d'agrafes pour instrument chirurgical |
| CN119745453A (zh) | 2019-05-31 | 2025-04-04 | 直观外科手术操作公司 | 用于外科器械的订合钉仓 |
| WO2021141971A1 (fr) | 2020-01-07 | 2021-07-15 | Intuitive Surgical Operations, Inc. | Instruments chirurgicaux pour appliquer de multiples agrafes |
| EP4274492B1 (fr) | 2021-01-08 | 2026-01-28 | Intuitive Surgical Operations, Inc. | Instrument chirurgical avec agrafes de sutures linéaires et en cordon de bourse |
| WO2022150215A1 (fr) | 2021-01-08 | 2022-07-14 | Intuitive Surgical Operations, Inc. | Instruments d'agrafage chirurgical |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100210975A1 (en) | 2009-01-21 | 2010-08-19 | SwimSense, LLC | Multi-state performance monitoring system |
| WO2010083562A1 (fr) * | 2009-01-22 | 2010-07-29 | National Ict Australia Limited | Détection d'activité |
| FR2943554B1 (fr) | 2009-03-31 | 2012-06-01 | Movea | Systeme et procede d'observation d'une activite de nage d'une personne |
| FR2964223B1 (fr) * | 2010-08-31 | 2016-04-01 | Commissariat Energie Atomique | Procede de configuration d'un dispositif de detection a capteur, programme d'ordinateur et dispositif adaptatif correspondants |
-
2012
- 2012-03-20 FR FR1252493A patent/FR2988501B1/fr not_active Expired - Fee Related
-
2013
- 2013-03-18 EP EP13715344.1A patent/EP2828797A1/fr not_active Withdrawn
- 2013-03-18 WO PCT/FR2013/050568 patent/WO2013140075A1/fr not_active Ceased
- 2013-03-18 US US14/386,549 patent/US9952043B2/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| None * |
| See also references of WO2013140075A1 * |
Also Published As
| Publication number | Publication date |
|---|---|
| US20150073746A1 (en) | 2015-03-12 |
| US9952043B2 (en) | 2018-04-24 |
| WO2013140075A1 (fr) | 2013-09-26 |
| FR2988501B1 (fr) | 2015-01-02 |
| FR2988501A1 (fr) | 2013-09-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2828797A1 (fr) | Dispositif et procede d'identification d'un mouvement cyclique, programme d'ordinateur correspondant | |
| EP2483758B1 (fr) | Systeme et procede de reconnaissance de gestes | |
| EP2423851B1 (fr) | Procédé de configuration d'un dispositif de détection à capteur, programme d'ordinateur et dispositif adaptatif correspondants | |
| WO2013030508A1 (fr) | Procédé de configuration d'un dispositif de détection à capteur, programme d'ordinateur et dispositif adaptatif correspondants | |
| FR2884007A1 (fr) | Procede d'identification de visages a partir d'images de visage, dispositif et programme d'ordinateur correspondants | |
| EP3770805B1 (fr) | Procede d'identification d'une personne dans une video, par une signature visuelle de cette personne, programme d'ordinateur et dispositif correspondants | |
| EP3252563B1 (fr) | Détermination d'un contexte de mobilité d'un utilisateur porteur d'un équipement muni de capteurs inertiels | |
| EP3937071A1 (fr) | Procédé d'assistance au suivi en temps réel d'au moins une personne sur des séquences d'images | |
| EP2767882B1 (fr) | Procédé de segmentation temporelle d'un geste instrumenté, dispositif, terminal et programme associés | |
| EP3770806B1 (fr) | Procede de surveillance video du franchissement d'une ligne par des personnes, programme d'ordinateur et dispositif correspondants | |
| EP3614306A1 (fr) | Procédé de localisation et d'identification de visage et de détermination de pose, à partir d'une vue tridimensionnelle | |
| EP2638492B1 (fr) | Dispositif de détection à capteur, procédé de détection et programme d'ordinateur correspondants | |
| EP3773161B1 (fr) | Dispositif informatique de détection de troubles du rythme cardiaque | |
| EP2881841B1 (fr) | Procédé de reconnaissance continue de gestes d'un utilisateur d'un terminal mobile préhensible muni d'un ensemble capteur de mouvement, et dispositif associé | |
| FR2998074A1 (fr) | Procede d'identification | |
| EP2988249A1 (fr) | Procédé de détermination, dans une image, d'au moins une zone susceptible de représenter au moins un doigt d'un individu | |
| EP4155987B1 (fr) | Procédé de modification d'une image contenant des données biométriques et dispositif associé | |
| EP3797509B1 (fr) | Traitement d'un bruit impulsionnel dans une séquence vidéo | |
| EP4292013A1 (fr) | Dispositif et procede de traitement de donnees videos pour detection du vivant | |
| EP4402648B1 (fr) | Procédé de détection d'un mouvement d'un corps | |
| EP4261777A1 (fr) | Procédé, programme d'ordinateur, dispositif et système de suivi d'un objet cible | |
| EP3022685B1 (fr) | Procede d'identification rapide d'un objet biometrique durant la phase d'acquisition et dispositif d'identification pour la mise en oeuvre de ce procede | |
| WO2018046868A1 (fr) | Procédé de reconnaissance de motif dans une pluralité de signaux |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| 17P | Request for examination filed |
Effective date: 20140916 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| AX | Request for extension of the european patent |
Extension state: BA ME |
|
| RIN1 | Information on inventor provided before grant (corrected) |
Inventor name: GRIS, FLORENCE Inventor name: JALLON, PIERRE |
|
| DAX | Request for extension of the european patent (deleted) | ||
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
| 17Q | First examination report despatched |
Effective date: 20180306 |
|
| GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: GRANT OF PATENT IS INTENDED |
|
| INTG | Intention to grant announced |
Effective date: 20201214 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
| 18D | Application deemed to be withdrawn |
Effective date: 20210427 |