Abstract: A system and a method for training a semantic segmentation model includes obtaining a plurality of sets of images each having an index z for visibility, iteratively training the model. Iteratively training the model includes (a) for each z above 1, obtaining preliminary semantic segmentation labels for each image of the set of images of index z?1 by applying the model to each image of the set of images of index z?1, (b) processing each preliminary semantic segmentation labels using semantic segmentation labels obtained using the model on a selected image of index 1, and obtaining processed semantic segmentation labels, (c) training the model using the set of images of index z?1 and the associated processed semantic segmentation labels, and (d) performing steps (a) to (c) for z+1.

Abstract: A system for producing a virtual image view for a vehicle is provided. The system includes one or more image capture means configured to capture image data in proximity to the vehicle, the image data being defined at least in part by first viewpoint parameters, and to provide an identifier identifying the respective one or more image capture means, storage means configured to store a plurality of virtualization records containing conversion information related to a virtualized viewpoint and a plurality of image capture means, and processing means.

Abstract: A system for generating a mask for object instances in an image is provided. The system includes a first module comprising a trained neural network and configured to input the image to the neural network, wherein the neural network is configured to generate: pixel offset vectors for the pixels of the object instance configured to point towards a unique center of an object instance, the pixel offset vectors thereby forming a cluster with a cluster distribution, and for each object instance an estimate of said cluster distribution defining a margin for determining which pixels belong to the object instance. A method for training a neural network map to be used for generating a mask for object instances in an image is also provided.

Abstract: A computer-implemented method for training a classifier (??), including: training a pretext model (??) to learn a pretext task, so as to minimize a distance between an output of a source sample via the pretext model (??) and an output of a corresponding transformed sample via the pretext model (??), the transformed sample being a sample obtained by applying a transformation (T) to the source sample; S20) determining a neighborhood (NXi) of samples (Xi) of a dataset (SD) in the embedding space; S30) training the classifier (??) to predict respective estimated probabilities ??j(Xi), j=1 . . . C, for a sample (Xi) to belong to respective clusters (Cj), by using a second training criterion which tends to: maximize a likelihood for a sample and its neighbors (Xj) of its neighborhood (Nxi) to belong to the same cluster; and force the samples to be distributed over several clusters.

Abstract: A system for image completion is disclosed. The system comprises a coordinate generation module configured to receive past frames and a present frame having a first field-of-view and to generate a set of coordinate maps, one for each of the received past frames; and a frame aggregation module configured to receive as input the past frames, the present frame, and the coordinate maps and to synthesize, based on said input, a present frame having a second field-of-view.

Abstract: A depth maps prediction system comprising a neural network (1000) configured to receive images (I) of a scene at successive time steps (t?1, t, t+1, . . . ) and comprising three sub-networks: an encoder (100), a ConvLSTM (200) and a decoder (300). The neural network (1000) is configured so that at each time step: a) the encoder sub-network (100) processes an image (I) and outputs a low resolution initial image representation (X); b) the CONVLSTM sub-network (200) processes the initial image representation (X), values for a previous time step (t?1) of an internal state (C(t?1)) and of an LSTM hidden variable data (H(t?1)) of the ConvLSTM sub-network, and outputs updated values of the internal state (C(t)) and of the LSTM hidden variable data (H(t)); and c) the decoder sub-network (300) inputs the LSTM output data (LOD) and generates a predicted dense depth map (D?) for the inputted image (I).

Abstract: A method for adjusting an information system of a mobile machine, the information system being configured to calculate 3D information relative to a scene in which the mobile machine is moving, the method including: acquiring at least a first image of the scene at a first time and a second image of the scene at a second time; detecting one or more scene features in the first image and the second image; matching the one or more scene features across the first image and the second image based upon detection of the one or more scene features; estimating an egomotion of the mobile machine based upon the matching of the one or more scene features across the first image and the second image; and adjusting the information system by taking into account the estimation of the egomotion of the mobile machine.

Abstract: A method for tracking a position of an object in a scene surrounding a mobile machine based upon information acquired from monocular images, includes: acquiring at least a first image at a first time and a second image at a second time, the first image and the second image each including image data corresponding to the object and a scene feature present in the scene surrounding the mobile machine; detecting the object in the first image and the second image; matching the scene feature across the first image and the second image; performing an estimation of an egomotion of the mobile machine based upon the scene feature matched across the first image and the second image; and predicting a position of the object taking into account the estimation of the egomotion of the mobile machine.

Abstract: A system and a method for training a model to be used for semantic segmentation of images, comprising: a—obtaining (S01) a first plurality of foggy images (101), b—training (S02) a classification model for estimating fog density, c—classifying (S03) a second plurality of images (101) having light fog, d—obtaining (S04) a third plurality of foggy images (103) having light fog, e—training (S05) a semantic segmentation model using the third plurality of foggy images, f—applying (S06) the semantic segmentation model to the second plurality of foggy (102) images to obtain semantic segmentations (102?), g—obtaining (S07) a fourth plurality of foggy images (104) having dense fog, h—training (S08) using the previously obtained foggy images.

Abstract: A system and a method for processing a plurality of images, each image of the plurality of images being acquired by a respective image acquisition module of a vehicle and each image acquisition module being oriented outwardly with respect to the vehicle, the method comprising: elaborating a bird's eye view image of surroundings of the vehicle using pixel values of pixels of at least one portion of each image of the plurality of images as pixel values of the bird's eye view image, and performing, on the bird's eye view image, a detection of at least one lane marked on a surface on which the vehicle is and visible on the bird's eye view image.

Abstract: A system and a method for processing an image so as to perform instance segmentation. The system/method includes: a—inputting (S1) the image (IMG) to a first neural network configured to output an affinity graph (AF), and b—inputting (S2), to a second neural network, the affinity graph and a predefined seed-map (SM), so as to determine whether other pixels belong to a same instance, and set at a first value the value of the other pixels determined as belonging to the same instance.

Abstract: A system for remote identification of users. The system uses deep learning techniques for authenticating a user from an identification document and using automated verification of identification documents. Identification documents may be authenticated by validating security features. The system may determine features expected in a valid identification document and determine whether those features are present, employing techniques, such as determining whether direction-sensitive features are present. Liveness of a user indicated by the identification document may be determined with a deep learning model trained for identification of facial spoofing attacks.

Abstract: A neural network, a system using this neural network and a method for training a neural network to output a description of the environment in the vicinity of at least one sound acquisition device on the basis of an audio signal acquired by the sound acquisition device, the method including: obtaining audio and image training signals of a scene showing an environment with objects generating sounds, obtaining a target description of the environment seen on the image training signal, inputting the audio training signal to the neural network so that the neural network outputs a training description of the environment, and comparing the target description of the environment with the training description of the environment.

Abstract: A system and a method for processing a plurality of images, each image of the plurality of images being acquired by a respective image acquisition module of a vehicle and each image acquisition module being oriented outwardly with respect to the vehicle, the method comprising: elaborating a bird's eye view image of surroundings of the vehicle using pixel values of pixels of at least one portion of each image of the plurality of images as pixel values of the bird's eye view image, and performing, on the bird's eye view image, a detection of at least one lane marked on a surface on which the vehicle is and visible on the bird's eye view image.

Abstract: A system for remote identification of users. The system uses deep learning techniques for authenticating a user from an identification document, using automated verification of identification documents and detection that a live person identified by the document is present. Liveness of a user indicated by the identification document may be determined with a deep learning model trained for identification of facial spoofing attacks. The deep learning model may be trained using training data extracted from facial feature locations of training images.

Abstract: A system and a method for training a semantic segmentation model includes obtaining a plurality of sets of images each having an index z for visibility, iteratively training the model. Iteratively training the model includes (a) for each z above 1, obtaining preliminary semantic segmentation labels for each image of the set of images of index z?1 by applying the model to each image of the set of images of index z?1, (b) processing each preliminary semantic segmentation labels using semantic segmentation labels obtained using the model on a selected image of index 1, and obtaining processed semantic segmentation labels, (c) training the model using the set of images of index z?1 and the associated processed semantic segmentation labels, and (d) performing steps (a) to (c) for z+1.

Abstract: A depth maps prediction system comprising a neural network (1000) configured to receive images (I) of a scene at successive time steps (t?1, t, t+1, . . . ) and comprising three sub-networks: an encoder (100), a ConvLSTM (200) and a decoder (300). The neural network (1000) is configured so that at each time step: a) the encoder sub-network (100) processes an image (I) and outputs a low resolution initial image representation (X); b) the CONVLSTM sub-network (200) processes the initial image representation (X), values for a previous time step (t?1) of an internal state (C(t?1)) and of an LSTM hidden variable data (H(t?1)) of the ConvLSTM sub-network, and outputs updated values of the internal state (C(t)) and of the LSTM hidden variable data (H(t)); and c) the decoder sub-network (300) inputs the LSTM output data (LOD) and generates a predicted dense depth map (D?) for the inputted image (I).

Abstract: A system for producing a virtual image view for a vehicle is provided. The system includes one or more image capture means configured to capture image data in proximity to the vehicle, the image data being defined at least in part by first viewpoint parameters, and to provide an identifier identifying the respective one or more image capture means, storage means configured to store a plurality of virtualization records containing conversion information related to a virtualized viewpoint and a plurality of image capture means, and processing means.

Abstract: A system for image completion is disclosed. The system comprises a coordinate generation module configured to receive past frames and a present frame having a first field-of-view and to generate a set of coordinate maps, one for each of the received past frames; and a frame aggregation module configured to receive as input the past frames, the present frame, and the coordinate maps and to synthesize, based on said input, a present frame having a second field-of-view.

Abstract: A system for generating a mask for object instances in an image is provided. The system includes a first module comprising a trained neural network and configured to input the image to the neural network, wherein the neural network is configured to generate: pixel offset vectors for the pixels of the object instance configured to point towards a unique center of an object instance, the pixel offset vectors thereby forming a cluster with a cluster distribution, and for each object instance an estimate of said cluster distribution defining a margin for determining which pixels belong to the object instance. A method for training a neural network map to be used for generating a mask for object instances in an image is also provided.