Abstract: This method for determining a skin colour of a region of interest of a user, comprises: —importing, from an image library (LIB) of a user computational device (APP), at least one image (IMk) comprising a representation of the region of interest of the user; —performing processing (TS) of said at least one image (IMk) with a machine-learning model (AI_TS) suitable for providing, for each imported image (IMk), a numerical value (ClrEstm_k) representative of the skin colour of the region of interest present in said at least None imported image (IMk); —performing evaluation (EVAL) of the skin colour of the region of interest on the basis of said one or more numerical values (ClrEstm_k) of each imported image.

Abstract: Device for forming by carding one or more fleece(s) of fibres comprising a main carding drum (1) supplied with raw fibres, one or more doffer cylinder(s) (3) capable of removing the fibres worked on the carding drum and an associated stripper roller to strip the fleece(s) of fibres formed on the or each doffer and deposit them on one or more output belts, the device also comprising at least one comb-cleaner flywheel (4) capable of interacting with at least one doffer cylinder to remove dust and impurities therefrom, characterised in that it/they is/are fitted with prevention means (3, 4) which are configured so as to prevent the dust and impurities removed from the comb by the cleaner flywheel from escaping in the direction of the associated stripper roller.

Abstract: A computer-implemented method of signal processing comprises providing images. The method comprises for each respective one of at least a subset of the images: applying a weakly-supervised learnt function, the weakly-supervised learnt function outputting respective couples each including a respective localization and one or more respective confidence scores, each confidence score representing a probability of instantiation of a respective object category at the respective localization. The method further comprises determining, based on the output of the weakly-supervised learnt function, one or more respective annotations, each annotation including a respective localization and a respective label representing instantiation a respective object category at the respective localization. The method further comprises forming a dataset including pieces of data, each piece of data including a respective image of the subset and at least a part of the one or more annotations determined for the respective image.

Abstract: The disclosure notably relates to a computer-implemented method for generating a 3D model representing a building. The method comprises providing a 2D floor plan representing a layout of the building. The method also comprises determining a semantic segmentation of the 2D floor plan. The method also comprises determining the 3D model based on the semantic segmentation. Such a method provides an improved solution for processing a 2D floor plan.

Abstract: The disclosure notably relates to a computer-implemented method for teaching a generative autoencoder. The generative autoencoder is configured to generate functional structures. A functional structure is a data structure representing a mechanical assembly of rigid parts and which includes a tree. Each leaf node represents a shape and positioning of a respective rigid part and a force exerted on the respective rigid part. Each non-leaf node with several children represents a mechanical link between sub-assemblies. Each sub-assembly is represented by a respective one of the several children. Each non-leaf node with a single child represents a duplication of the sub-assembly represented by the single child. The method includes obtaining a dataset including functional structures. The method further comprises teaching the generative autoencoder on the dataset. This constitutes an improved method for teaching a generative autoencoder configured for generating functional structures.

Abstract: A device is provided for determining an RF coverage quality of a wireless mesh network at a given location. The device includes a communication module, a processing module, and a visual interface. The communication module is configured to communicate with nearby neighbours in the wireless mesh network. The processing module is configured to detect and calculate good neighbours having an RF signal strength above a first predetermined threshold, and a tickle success rate above a second predetermined threshold. Methods to install and validate the installation of a wireless electric meter and a passive antenna are also provided.

Abstract: A computer-implemented method of signal processing comprises providing images. The method comprises for each respective one of at least a subset of the images: applying a weakly-supervised learnt function, the weakly-supervised learnt function outputting respective couples each including a respective localization and one or more respective confidence scores, each confidence score representing a probability of instantiation of a respective object category at the respective localization. The method further comprises determining, based on the output of the weakly-supervised learnt function, one or more respective annotations, each annotation including a respective localization and a respective label representing instantiation a respective object category at the respective localization. The method further comprises forming a dataset including pieces of data, each piece of data including a respective image of the subset and at least a part of the one or more annotations determined for the respective image.

Abstract: A computer-implemented method of signal processing comprises providing images. The method comprises for each respective one of at least a subset of the images: applying a weakly-supervised learnt function, the weakly-supervised learnt function outputting respective couples each including a respective localization and one or more respective confidence scores, each confidence score representing a probability of instantiation of a respective object category at the respective localization. The method further comprises determining, based on the output of the weakly-supervised learnt function, one or more respective annotations, each annotation including a respective localization and a respective label representing instantiation a respective object category at the respective localization. The method further comprises forming a dataset including pieces of data, each piece of data including a respective image of the subset and at least a part of the one or more annotations determined for the respective image.

Abstract: The disclosure notably relates to a computer-implemented method for teaching a generative autoencoder. The generative autoencoder is configured to generate functional structures. A functional structure is a data structure representing a mechanical assembly of rigid parts and which includes a tree. Each leaf node represents a shape and positioning of a respective rigid part and a force exerted on the respective rigid part. Each non-leaf node with several children represents a mechanical link between sub-assemblies. Each sub-assembly is represented by a respective one of the several children. Each non-leaf node with a single child represents a duplication of the sub-assembly represented by the single child. The method includes obtaining a dataset including functional structures. The method further comprises teaching the generative autoencoder on the dataset. This constitutes an improved method for teaching a generative autoencoder configured for generating functional structures.

Abstract: A device is provided for determining an RF coverage quality of a wireless mesh network at a given location. The device includes a communication module, a processing module, and a visual interface. The communication module is configured to communicate with nearby neighbours in the wireless mesh network. The processing module is configured to detect and calculate good neighbours having an RF signal strength above a first predetermined threshold, and a tickle success rate above a second predetermined threshold. Methods to install and validate the installation of a wireless electric meter and a passive antenna are also provided.

Abstract: Cognitive training handle designed to be connected by a longitudinal connecting means to a connection means on a ball, said handle also comprising a means for adjusting the length of said longitudinal connecting means.

Abstract: The disclosure notably relates to a computer-implemented method for determining a function configured to determine a semantic segmentation of a 2D floor plan representing a layout of a building. The method comprises providing a dataset comprising 2D floor plans each associated to a respective semantic segmentation. The method also comprises learning the function based on the dataset. Such a method provides an improved solution for processing a 2D floor plan.

Abstract: The disclosure notably relates to a computer-implemented method for generating a 3D model representing a building. The method comprises providing a 2D floor plan representing a layout of the building. The method also comprises determining a semantic segmentation of the 2D floor plan. The method also comprises determining the 3D model based on the semantic segmentation. Such a method provides an improved solution for processing a 2D floor plan.

Abstract: A computer-implemented method of signal processing comprises providing images. The method comprises for each respective one of at least a subset of the images: applying a weakly-supervised learnt function, the weakly-supervised learnt function outputting respective couples each including a respective localization and one or more respective confidence scores, each confidence score representing a probability of instantiation of a respective object category at the respective localization. The method further comprises determining, based on the output of the weakly-supervised learnt function, one or more respective annotations, each annotation including a respective localization and a respective label representing instantiation a respective object category at the respective localization. The method further comprises forming a dataset including pieces of data, each piece of data including a respective image of the subset and at least a part of the one or more annotations determined for the respective image.

Abstract: Cognitive training handle designed to be connected by a longitudinal connecting means to a connection means on a ball, said handle also comprising a means for adjusting the length of said longitudinal connecting means.

Abstract: According to the invention, a time synchronization of phase between measurement devices that do not share a same clock for their respective sampling of the signals is carried out by a time tagging of samples of the signals in time blocks followed by an adjustment of the phase values of components of interest of the signals in the regrouped time blocks so that the values refer to common time references between the measurement devices. The tagging is carried out with a synchronization signal available to the measurement devices, completed with count values provided by a counter operated by a reference clock for each measurement device.

Abstract: A method and apparatus for signal acquisition in an OFDM receiver relies on a preamble training sequence to synchronize the receiver in time (e.g. determining the start of a frame) and in frequency (carrier frequency offset compensation). The preamble training sequence has a periodic structure and the method and apparatus perform a cross-correlation technique using a matched filter to achieve time synchronization and/or frequency synchronization and/or channel estimation, the latter being especially useful in multi-antenna receivers for diversity combining purposes. Many advantages derive from performing at least two and preferably all three operations jointly, in terms of latency, hardware complexity, and length of training sequence required to achieve satisfactory convergence on all counts. The periodicity of the training sequence is exploited to reduce considerably the main filter complexity and optionally dynamically adjust carrier offset compensation throughout the filtering process.

Abstract: An unhindered polyol is used to react with an epoxidized soyate to make epoxidized soyate diester in the presence of a catalyst. The unhindered polyol can be 1,3-propanediol or any polyol having four or more carbon atoms with no two adjacent carbon atoms having hydroxyl functionality. Preferably, a combination of catalysts is used to promote the transesterification reaction of the epoxidized soyate with the unhindered polyol to yield a high percentage of epoxidized soyate diester with epoxy functionality retained. The primary catalyst is a metallic hydroxide, and the secondary catalyst is a titanate. Bioderived epoxidized soyate diester plasticizers useful for thermoplastics and thermosets result.

Abstract: A crosslapper receives a card web and folds it into a lap intended to be needle-punched or consolidated by other ways. The web includes zones which are more condensed, having a spectrum of orientation of fibers with a component parallel to the width of the web, alternating with less condensed zones having a longitudinal unidirectional spectrum of orientations. The zones which are less condensed are used to form the edge zones of the lap. The result is that the lap has different respective spectra of orientation which pre-compensate for the unwanted changes produced by the needle-punching or other consolidation which follows. A needle-punched lap is obtained having a uniform MD/CD ratio (relationship between longitudinal and respectively transverse tensile strengths) or having a sought profile of the MD/CD ratio across the width of the lap.

Abstract: According to the invention, a time synchronization of phase between measurement devices that do not share a same clock for their respective sampling of the signals is carried out by a time tagging of samples of the signals in time blocks followed by an adjustment of the phase values of components of interest of the signals in the regrouped time blocks so that the values refer to common time references between the measurement devices. The tagging is carried out with a synchronization signal available to the measurement devices, completed with count values provided by a counter operated by a reference clock for each measurement device.