Abstract: A computer-implemented method for determining a machine-learning function configured for taking an input 3D scene and for outputting one or more camera viewpoints each for generating a respective 2D rendering of the 3D scene. The method includes obtaining a library having 3D scenes. The method includes, based on the library, forming a first dataset for training a first neural network configured for outputting a camera position and forming a second dataset for training a second neural network configured for outputting a camera orientation. The method includes training the first neural network based on the first dataset and training the second neural network based on the second dataset. Each camera viewpoint outputted by the machine-learning function includes a camera position and a camera orientation. Such a method forms an improved solution for outputting one or more camera viewpoints of a 3D scene.

Abstract: A computer-implemented method of applying a machine-learning function preconfigured for taking an input 3D layout and a given noise level, and for predicting an output 3D layout. The function is preconfigured with a conditioning drop-out with respect to at least one layout parameter. The method further comprises obtaining a set of conditioning inputs and, for each conditioning input, determining one or more conditioning candidate 3D layouts and determining a plurality of perturbed conditioning candidate 3D layouts. The method further includes, applying the preconfigured function to each perturbed conditioning candidate, in which the one layout parameter is dropped out, computing reconstruction errors, and averaging the reconstruction errors, thereby obtaining a score. This forms an improved solution for predicting 3D layouts.

Abstract: A computer-implemented method of machine-learning. The method includes obtaining a dataset of ground truth 3D layouts. The machine-learning method further comprises obtaining a probability distribution of noise levels. The machine-learning method also comprises, for each ground truth 3D layout, obtaining a respective perturbed 3D layout. The machine-learning method moreover comprises training a function. The function is configured for taking an input 3D layout and a given noise level, and for predicting an output 3D layout. The training is performed over the dataset based on a loss which penalizes a dissimilarity metric between each ground truth 3D layout and a respective predicted 3D layout.

Abstract: A computer-implemented method for generating a 2D image of a 3D scene. The method comprises obtaining arrangement data comprising a layout of the 3D scene and at least one conditioning signal. Each conditioning signal has a type among a predetermined set of at least two types. The method comprises applying a machine-learning function to the obtained arrangement data and viewpoint. The function comprises a scene encoder and a generative image model. The scene encoder takes as input the obtained arrangement data and viewpoint and outputting a scene encoding tensor. The generative image model takes as input the scene encoding tensor outputted by the scene encoder and outputting the generated 2D image. Such a generating method forms an improved solution for controllably generating a 2D image of a 3D scene.

Abstract: A computer-implemented method for generating a plurality of 2D images of a 3D scene. The method comprises obtaining arrangement data and a machine-learning model configured for generating a 2D image. The method comprises generating a plurality of first 2D images of the 3D scene each having a respective viewpoint. The method comprises, for each generated first 2D image, computing a first latent vector. The method comprises, for each generated first 2D image, computing a second latent vector as a weighted combination of the computed first latent vectors. The method comprises generating a plurality of second 2D images of the 3D scene by applying, for each given viewpoint of the first 2D images, the model to the computed second latent vector. Such a generating method forms an improved solution for generating a plurality of 2D images of a given 3D scene.

Abstract: A computer-implemented method of machine-learning for learning a neural network that encodes a super-point of a 3D point cloud into a latent vector. The method including obtaining a dataset of super-points. Each super-point is a set of points of a 3D point cloud. The set of points represents at least a part of an object. The method further includes learning the neural network based on the dataset of super-points. The learning includes minimizing a loss. The loss penalizes a disparity between two super-points. This constitutes improved machine-learning for 3D object detection.

Abstract: A computer-implemented method for machine learning a function configured to take as input a 3D point cloud of a real scene and to output localized representations each of a respective object of the real scene and, for each respective object, a class of the respective object among a predetermined set of classes. The method comprises obtaining a dataset of 3D point clouds annotated, for each 3D point cloud, with localized representations each of a respective object and, for each respective object, with a class of the respective object among the predetermined set of classes. The method includes training the function based on the obtained dataset. The predetermined set of classes comprises a plurality of semantic classes and a plurality of geometric classes. Such a method forms an improved solution for 3D scene understanding.

Abstract: A computer-implemented method for machine-learning a function that generates a 2D image of a 3D scene. The function includes a scene encoder and a generative image model. The scene encoder takes as input a layout of the 3D scene and a viewpoint and outputs a scene encoding tensor. The generative image model takes as input the scene encoding tensor outputted by the scene encoder and outputs the generated 2D image. The machine-learning method includes obtaining a dataset comprising 2D images and corresponding layouts and viewpoints of 3D scenes. The machine-learning method includes training the function based on the obtained dataset. Such a machine-learning method forms an improved solution for generating a 2D image of a 3D scene.

Abstract: A computer-implemented method for determining a machine-learning function configured for taking an input 3D scene and for outputting one or more camera viewpoints each for generating a respective 2D rendering of the 3D scene. The method includes obtaining a library having 3D scenes. The method includes, based on the library, forming a first dataset for training a first neural network configured for outputting a camera position and forming a second dataset for training a second neural network configured for outputting a camera orientation. The method includes training the first neural network based on the first dataset and training the second neural network based on the second dataset. Each camera viewpoint outputted by the machine-learning function includes a camera position and a camera orientation. Such a method forms an improved solution for outputting one or more camera viewpoints of a 3D scene.

Abstract: A computer-implemented method for generating a training dataset. The training dataset includes training patterns each including a 3D point cloud of a respective travelable environment. The generating method includes, for each 3D point cloud, obtaining a 3D surface representation of the respective travelable environment, determining a traveling path inside the respective travelable environment, and, generating a virtual scan of the respective travelable environment along the traveling path, thereby obtaining the 3D point cloud. Such a method forms an improved solution for generating a training dataset of 3D point clouds.

Abstract: A computer-implemented method of machine-learning including obtaining a dataset of 3D point clouds. Each 3D point cloud includes at least one object. Each 3D point cloud is equipped with a specification of one or more graphical user-interactions each representing a respective selection operation of a same object in the 3D point cloud. The method further includes teaching, based on the dataset, a neural network configured for segmenting an input 3D point cloud including an object. The segmenting is based on the input 3D point cloud and on a specification of one or more input graphical user-interactions each representing a respective selection operation of the object in the 3D point cloud.

Abstract: A computer-implemented method for automatically furnishing a 3D room based on user preferences including obtaining at least one spatial relations graph of a virtual 3D room and a set of user preferences, converting the set of user preferences into a set of target parameters, computing, for each spatial relations graph: a set of Key Performance Indicator values, and a KPI distance, automatically selecting at least one most promising spatial relations graph, instantiating the most promising spatial relations graph into the 3D room to be furnished, displaying the furnished virtual 3D room proposal to the user, receiving an update of the user preferences, and reiterating until a stopping criterion is fulfilled.

Abstract: A computer-implemented method for automatically providing at least a template of a furnished virtual 3D room including 3D elements, including, for each furnished virtual 3D room: automatically extracting a spatial relations graph based on spatial relations between the 3D elements of the 3D room, said 3D elements including 3D architectural elements and 3D furnishing objects located in the furnished virtual 3D room, automatically extracting at least one zone from the 3D room based on the spatial relations graph, and extracting a set of constraints about a relative arrangement of said zone with respect to the room architecture or with respect to other zones of the 3D room, a zone being defined by a local spatial arrangement of at least one 3D furnishing object, and storing a template of the furnished virtual 3D room, said template including said zone and said set of constraints.

Abstract: A computer-implemented method of machine-learning for learning a neural network that encodes a super-point of a 3D point cloud into a latent vector. The method including obtaining a dataset of super-points. Each super-point is a set of points of a 3D point cloud. The set of points represents at least a part of an object. The method further includes learning the neural network based on the dataset of super-points. The learning includes minimizing a loss. The loss penalizes a disparity between two super-points. This constitutes improved machine-learning for 3D object detection.

Abstract: A computer-implemented method of machine-learning including obtaining a dataset of 3D point clouds. Each 3D point cloud includes at least one object. Each 3D point cloud is equipped with a specification of one or more graphical user-interactions each representing a respective selection operation of a same object in the 3D point cloud. The method further includes teaching, based on the dataset, a neural network configured for segmenting an input 3D point cloud including an object. The segmenting is based on the input 3D point cloud and on a specification of one or more input graphical user-interactions each representing a respective selection operation of the object in the 3D point cloud.