Abstract: A computer-implemented method to perform image-to-image translation. The method can include obtaining one or more machine-learned generator models. The one or more machine-learned generator models can be configured to receive an input image and a user-specified conditioning vector that parameterizes one or more desired values for one or more defined characteristics of an output image. The one or more machine-learned generator models can be configured to perform, based at least in part on the user-specified conditioning vector, one or more transformations on the input image to generate the output image with the one or more desired values for the one or more defined characteristics. The method can include receiving the input image and the user-specified conditioning vector. The method can include generating, using the machine-learned generator model, an output image having the one or more desired values for the one or more characteristics.

Abstract: A computer-implemented method to perform image-to-image translation. The method can include obtaining one or more machine-learned generator models. The one or more machine-learned generator models can be configured to receive an input image and a user-specified conditioning vector that parameterizes one or more desired values for one or more defined characteristics of an output image. The one or more machine-learned generator models can be configured to perform, based at least in part on the user-specified conditioning vector, one or more transformations on the input image to generate the output image with the one or more desired values for the one or more defined characteristics. The method can include receiving the input image and the user-specified conditioning vector. The method can include generating, using the machine-learned generator model, an output image having the one or more desired values for the one or more characteristics.

Abstract: Generally, the present disclosure is directed to systems and methods that leverage batch normalization statistics as a way to generalize across domains In particular, example implementations of the present disclosure can generate different representations for different domains by collecting independent batch normalization statistics, which can then be used to map between domains in a shared latent space. At test or inference time, samples from an unknown test or target domain can be projected into the same shared latent space. The domain of the target sample can therefore be expressed as a linear combination of the known ones, with the combination between weighted based on respective distances between batch normalization statistics in the latent space. This same mapping strategy can be applied at both training and test time to learn both a latent representation and a powerful but light-weight ensemble model that operates within such latent space.

Abstract: A computer-implemented method to perform image-to-image translation. The method can include obtaining one or more machine-learned generator models. The one or more machine-learned generator models can be configured to receive an input image and a user-specified conditioning vector that parameterizes one or more desired values for one or more defined characteristics of an output image. The one or more machine-learned generator models can be configured to perform, based at least in part on the user-specified conditioning vector, one or more transformations on the input image to generate the output image with the one or more desired values for the one or more defined characteristics. The method can include receiving the input image and the user-specified conditioning vector. The method can include generating, using the machine-learned generator model, an output image having the one or more desired values for the one or more characteristics.

Abstract: Example embodiments allow for fast, efficient detection and pose estimation of objects based on point clouds, depth images/maps, or other depth information about a scene that may contain the objects. Embodiments include translating and rotating the depth image to bring individual points of the depth image to a standard orientation and location so as to improve performance when an object is near the periphery of the field of view. Some disclosed embodiments include applying a random forest to perform pose estimation. By using the decision trees or other fast methods, it can be advantageous to perform pose estimation a plurality of times prior to identifying whether a particular object is actually present in a scene. Prospective pose estimates can be combined with models of the objects in order to evaluate whether the object is present in the scene.

Abstract: A system and a method for processing an image include inputting the image to a neural network configured to: obtain a plurality of feature maps, each feature map having a respective resolution and a respective depth, perform a classification on each feature map to deliver, for each feature map: the type of at least one object visible on the image, the position and shape in the image of at least one two-dimensional bounding box surrounding the at least one object, at least one possible viewpoint for the at least one object, at least one possible in-plane rotation for the at least one object.

Abstract: A system and a method for processing an image include inputting the image to a neural network configured to: obtain a plurality of feature maps, each feature map having a respective resolution and a respective depth, perform a classification on each feature map to deliver, for each feature map: the type of at least one object visible on the image, the position and shape in the image of at least one two-dimensional bounding box surrounding the at least one object, a plurality of rotation hypotheses for the at least one object.

Abstract: A system and a method for processing an image include inputting the image to a neural network configured to: obtain a plurality of feature maps, each feature map having a respective resolution and a respective depth, perform a classification on each feature map to deliver, for each feature map: the type of at least one object visible on the image, the position and shape in the image of at least one two-dimensional bounding box surrounding the at least one object, at least one possible viewpoint for the at least one object, at least one possible in-plane rotation for the at least one object.

Abstract: A computer-implemented method to perform image-to-image translation. The method can include obtaining one or more machine-learned generator models. The one or more machine-learned generator models can be configured to receive an input image and a user-specified conditioning vector that parameterizes one or more desired values for one or more defined characteristics of an output image. The one or more machine-learned generator models can be configured to perform, based at least in part on the user-specified conditioning vector, one or more transformations on the input image to generate the output image with the one or more desired values for the one or more defined characteristics. The method can include receiving the input image and the user-specified conditioning vector. The method can include generating, using the machine-learned generator model, an output image having the one or more desired values for the one or more characteristics.

Abstract: During a description technique, a local descriptor for an object may be generated by computing a 2-dimensional histogram of pairs of angles between pairs of line segments that are aligned with edge pixels associated with the object. The pairs of line segments belong to a subset of k neighboring or proximate line segments. Moreover, this 2D histogram may represent the relative displacement and the relative orientations of the pairs of line segments in the subset as weights in bins or cells defined by angular quantization values, and the 2D histogram may exclude lengths of the line segments. Subsequently, the generated 2D histogram may be compared to predefined sets of descriptors in a model library that are associated with a set of objects, and the object may be identified as one of the set of objects based on a group of match scores determined in the comparisons.

Abstract: During a description technique, a local descriptor for an object may be generated by computing a 2-dimensional histogram of pairs of angles between pairs of line segments that are aligned with edge pixels associated with the object. The pairs of line segments belong to a subset of k neighboring or proximate line segments. Moreover, this 2D histogram may represent the relative displacement and the relative orientations of the pairs of line segments in the subset as weights in bins or cells defined by angular quantization values, and the 2D histogram may exclude lengths of the line segments. Subsequently, the generated 2D histogram may be compared to predefined sets of descriptors in a model library that are associated with a set of objects, and the object may be identified as one of the set of objects based on a group of match scores determined in the comparisons.

Abstract: During a description technique (100), a local descriptor for an object (300) is generated (122) by computing a 2-dimensional histogram (600) of pairs of angles (514, 516) between pairs of line segments (510, 512) that are aligned with edge pixels associated with the object (300). The pairs of line segments (510, 512) belong to a subset of k neighboring or proximate line segments (310). Moreover, this 2D histogram (600) may represent the relative displacement and the relative orientations of the pairs of line segments (510, 512) in the subset as weights in bins or cells defined by angular quantization values, and the 2D histogram (600 may exclude lengths of the line segments. Subsequently, the generated 2D histogram (600) may be compared (210) to predefined sets of descriptors in a model library that are associated with a set of objects, and the object may be identified (212) as one of the set of objects based on a group of match scores determined in the comparisons.

Abstract: During a description technique (100), a local descriptor for an object (300) is generated (122) by computing a 2-dimensional histogram (600) of pairs of angles (514, 516) between pairs of line segments (510, 512) that are aligned with edge pixels associated with the object (300). The pairs of line segments (510, 512) belong to a subset of k neighboring or proximate line segments (310). Moreover, this 2D histogram (600) may represent the relative displacement and the relative orientations of the pairs of line segments (510, 512) in the subset as weights in bins or cells defined by angular quantization values, and the 2D histogram (600 may exclude lengths of the line segments. Subsequently, the generated 2D histogram (600) may be compared (210) to predefined sets of descriptors in a model library that are associated with a set of objects, and the object may be identified (212) as one of the set of objects based on a group of match scores determined in the comparisons.

Abstract: Disclosed embodiments are directed to methods, systems, and circuits of generating compact descriptors for transmission over a communications network. A method according to one embodiment includes receiving an uncompressed descriptor, performing zero-thresholding on the uncompressed descriptor to generate a zero-threshold-delimited descriptor, quantizing the zero-threshold-delimited descriptor to generate a quantized descriptor, and coding the quantized descriptor to generate a compact descriptor for transmission over a communications network. The uncompressed and compact descriptors may be 3D descriptors, such as where the uncompressed descriptor is a SHOT descriptor. The operation of coding can be ZeroFlag coding, ExpGolomb coding, or Arithmetic coding, for example.