Abstract: The invention in particular concerns a method and a device for estimating a level of use of a communication network linking a server and at least one client, said communication network being used to transmit a data stream at a transmission rate corresponding to the temporal frequency of said data stream, said data stream comprising at least one item of information characterizing a theoretical rate dependent on said transmission rate of said data stream. After having obtained (121) said theoretical rate of said data stream, the real reception rate of at least one part of said data stream is evaluated (109). Said theoretical rate and said evaluated rate are then compared (113) to estimate said level of use of said communication network. The level of subscription of a client to multicast sessions may be adapted according to the level of use of the communication network.

Abstract: The method for transmitting video data encoded in a scalable video stream representing said video data in a plurality of layers, comprises, for at least one layer: a step of obtaining (317) pre-encoded macroblocks of said scalable video stream representing said video data in a plurality of layers, a step of grouping (319) macroblocks, used for inter layer prediction, of the pre-encoded macroblocks obtained by the obtaining step, into first groups of macroblocks, a step of grouping (319) macroblocks, not used for inter layer prediction, of the pre-encoded macroblocks obtained by the obtaining step, into second groups of macroblocks, a step of estimating a loss variability over a network, a step of selecting (331) groups of macroblocks from the first and second groups of macroblocks based on the loss variability estimated by the estimating step and a step of transmitting (333) the groups of macroblocks selected by the selecting step over the network.

Abstract: Video coding in which at least two bandwidth values are obtained. A base layer is coded so that a rate of the coded base layer data is less than or equal to the lowest bandwidth. For at least one bandwidth greater than the lowest bandwidth, at least one enhancement layer is coded so that the sum of a rate of the data of the coded enhancement layer and a rate of a subset of data of a reference layer serving as a reference for interlayer prediction of the enhancement layer is less than the bandwidth greater than the lowest bandwidth.

Abstract: The reception method is implemented in a receiver apparatus receiving images of a video sequence having undergone losses during an exchange of data with a sender apparatus connected to said receiver apparatus via a communication network. The receiver apparatus comprises at least a first module implementing a transport layer and a second module implementing an application layer, the application layer being subdivided into a first sub-module implementing the decoding of the video and a second sub-module implementing the display of said video. The method comprises the following steps: in the transport layer, transmitting a data loss detection signal intended for the application layer, as soon as the transport layer detects a loss of data; and in the application layer, implementing a loss concealment mechanism, on reception of said loss detection signal and without waiting for said application layer itself to detect said data loss so detected.

Abstract: The invention relates to a method and to a device for determining the value of a delay to be applied between sending a first dataset and sending a second dataset, the data being representative of a sequence of coded images, the datasets consisting of data subsets, the coded images being coded according to a first level of resolution and at least a second level of resolution higher than the first level of resolution, the data subsets containing data of a first level of resolution. According to the invention: first and second datasets are obtained (E506); and the value of the delay is determined (E507), the value of the delay being dependent on the number of decoded images that can be formed from the data subsets of the first dataset which contain data of the second or of a second level of resolution.

Abstract: A video signal pre-coded using hierarchical coding is transmitted over a communication network, the signal being constituted by data distributed into several hierarchical levels. More specifically, a bandwidth of the communication network is periodically estimated over evaluation periods. A set of hierarchical levels is selected for a current evaluation period, where a transmission rate of the data of the selected hierarchical levels is compatible with the bandwidth estimated for a current evaluation period. A change tendency of the bandwidth is estimated for a following evaluation period. Additional data of the video signal to transmit is then determined depending on the estimated change tendency of the bandwidth, and the data of the selected hierarchical levels and the determined additional data is transmitted during the current evaluation period.

Abstract: A method of coding a digital image which includes plural macroblocks of which one is designated as a “current” macroblock. Motion estimation is applied to the current macroblock and to at least one image designated as “reference”, to obtain one or more first predictors in each reference image. A second predictor is obtained from a spatial scalability level lower than the spatial scalability level of the current macroblock. At least one of the first predictors is transformed from the spatial domain to the frequential domain. At least one low spatial frequency coefficient is predicted on the basis of the second predictor, and at least one high spatial frequency coefficient is predicted on the basis of at least one high frequency coefficient of the transformed first predictor.

Abstract: The present invention concerns a method of coding a plurality of digital image signals in a bitstream, said signals each corresponding to a view of a scene, the said coding method comprising the use of a pre-determined predictive coding structure, the views being associated with input views in the coding structure. The coding method according to the invention comprises, for at least one pair of views of the scene, steps of determining a spatial intersection zone (E44) and of calculating a value representing the similarity between the signals corresponding to said views (E45), according to at least some of the signals contained in said spatial intersection zone. These steps are followed by a step of associating (E46) at least one view with an input view of the predictive coding structure according to at least one calculated similarity value. The invention applies both to fixed image signals and to image sequences.

Abstract: The method of transmitting a video stream over a network between a transmission device and at least one reception device comprises: —a step (502) of encoding so-called “primary” data of the video stream according to a first type of encoding, —a step (516, 518) of obtaining so-called “secondary” video data, dependent on the primary data, by the simulation of transmission errors potentially suffered by the video stream and at least one method of masking losses due to said transmission errors able to be implemented by a reception device able to decode the primary video stream encoded according to the first type of encoding, —a step (520, 522) of encoding secondary data according to a second type of encoding different from the first type of encoding, and —a step of transmitting, by means of the network, primary data encoded according to the first type of encoding and at least some of the secondary data encoded with the second type of encoding.

Abstract: For transmitting a sequence of video images on a network between a server and a client, that sequence being coded according to a hybrid predictive coding mode and comprising a plurality of images, each image being broken down into a plurality of macroblocks of pixels: the plurality of macroblocks is classified (E212) into at least two groups of macroblocks according to a predetermined criterion; these groups of macroblocks are completed (E216, E226) by additional macroblocks; at least two scalable coding levels are created (E214, E224) from these groups of macroblocks; unequal protection against transmission errors is applied (E218) to the groups of macroblocks based on the predetermined criterion; and the scalable coding levels are transmitted (E220) on the network.

Abstract: The invention relates to a method and a device for transmitting a current image of a sequence of digital images between a server device and a client device which are connected by a communication network. Each image of the sequence is encoded in a multi-resolution format, each resolution comprising a set of non independent elementary entities. The method of the invention comprises detecting motion, carried out on spatial entities defined with respect to said elementary entities, between the current image and a reference image, then determining a first set of elementary entities comprising the elementary entities relating to said at least one moving spatial entity and determining a second set of elementary entities which are necessary for reconstructing said moving spatial entities. Finally, at least part of each of the elementary entities of said first and second sets is sent to the client device.

Abstract: A video sequence is coded in the form of a base level and an refinement level. A group of pictures of the base level comprises intra coded or mono-predicted end images (I,P) and bi-predicted images (B1, B2, B3), the refinement level comprises at least one image (E2, E4, . . . , E16) with no temporal correspondent in the base level. Use is made of predictive coding of said image with no temporal correspondent (E2, E4, . . . , E16) by making reference to a plurality of images, each image of said plurality being an image of the refinement level with no temporal correspondent in the base level (E4, E8, E12, E14) or an image of the refinement level (E1, E17) corresponding temporally to an end image (I1P).

Abstract: The invention concerns methods and devices for concealing artefacts induced by transmission errors in a video sequence transmitted over a network between a server and a client making it possible to avoid abrupt resynchronization following a transmission error. When an error is detected, a video stream is created from a first video stream not containing any error and data representing artefacts induced by that error. The influence of the data representing the artefacts diminishes with time such that the created video stream converges towards the video stream not containing any error. The creation of the video stream in which the effects due to a transmission error are progressively removed may be carried out by the server or by the client.

Abstract: The method of transmitting data packets associated with an importance level comprises a step (305) of estimating available rate on the network. In case a variation is detected in estimated available rate greater than a first predetermined value, determination is made (309-317) of whether there is an increase in the round trip time on said network that is greater than a second predetermined value. If yes, the data packets are transmitted (319-325) at a rate lower than the estimated available rate. In embodiments, in case a variation is detected in estimated available rate greater than a first predetermined value and before the step of detecting an increase in the round trip time, a step (307) is carried out of transmitting packets of lower importance level than the average of said importance levels by using the current estimated available rate and the round trip time is determined on the basis of the sending of those packets.

Abstract: A method of reconstructing a sequence of video data in a receiving device after the video sequence has been transmitted in encoded form over a network, the encoded video sequence comprising containers of encoded video data, wherein the method includes the following steps: determining (809) containers of encoded video data affected by data losses during their transmission over the network, determining (813) the time that would be necessary for decoding said containers of encoded video data affected by data losses if they had not been affected by data losses, allocating (819, 820, 821) the time thus determined to reconstructing video data of the sequence.

Abstract: This method of transmitting coded data, representing a digital video, from a server to a plurality of clients, at least one subset of the clients using congestion control algorithms, uses a plurality of signals for passing from a coded representation of the video at a given quality level to a coded representation at least one quality level different from the given level. The sending (E618) of at least one of the passage signals at least to the subset of clients depends on information representing the behavior of the congestion control algorithms used by the subset of clients.

Abstract: The invention relates to a method of processing a multimedia data stream coded according to a plurality of hierarchical levels and transmitted over a communication network with loss, the hierarchy levels being ordered so that a so-called higher hierarchy level is coded in a way that is dependent on at least one so-called lower hierarchy level, at least one portion of multimedia data of a lower hierarchy level having suffered losses, the method comprising a step (515) of error concealment in said at least one portion of data having suffered losses, characterized in that it comprises a processing step applied to the error concealment in said at least one portion of data of the lower hierarchy level and involving data of at least one higher hierarchy level dependent on the lower hierarchy level.

Abstract: Selecting a method from a set of methods of transcoding digital video data, the data being transferable between at least two communication apparatuses via a communication network, including, for each transcoding method, obtaining a distortion value from a mathematical model for establishing the distortion taking into account the transcoding method, the distortion model being applied to information representing the content of the signal represented by the digital data to be transcoded, and selecting a transcoding method as a function of the different distortion values obtained for the set of transcoding methods, the set of transcoding methods including at least two different types of transcoding methods, each type of transcoding method being selected from at least one of a plurality of methods of reducing a spatial resolution, a plurality of a re-quantization methods, and a plurality of methods of reducing a frame rate.

Abstract: The invention relates to a method for the continuous transmission of a video from a server to a number of client machines over a communication network, comprising the following steps: obtaining requests from client machines, at least some of the requests obtained each specifying at least one spatial zone of interest in one or more images of the video, a spatial zone representing a set of digital data, processing the requests obtained so as to determine one or more intersections defining common data between the spatial zones of interest of the video, creating a multipoint network session for each intersection thus determined.

Abstract: To identify data losses in a video sequence transmitted between a server and at least one client over a communication network, the sequence comprising a plurality of video data containers coded using scalable video coding employing a predetermined number of hierarchical levels, the video data containers being transmitted over the network via a series of data transport packets: a process (311) is carried out of detecting losses and of locating the video data affected by those losses, without decoding the video data, by combining information coming from transport packets and information coming from video data container headers. Application for improving decoding on receiving a video sequence.