Abstract: Improved methods are provided for generating, via a noise-diffusion iterative process, depth maps or optical flow maps from input images. Also provided are improved methods for training the machine learning model(s) employed in the iterative process and for augmenting die set of training data used to train such models. By translating the depth or optical flow map prediction process into the noise diffusion context, improved performance with respect to compute cost, training data, requirements, model size, and output quality are obtained. Additionally, the noise diffusion context allows models trained as described herein to generate maps de novo from target color images and/or to begin from initial ‘guess’ maps (e.g., noisy maps, maps containing holes) when generating improved output maps, natively incorporating the imperfect prior information represented by such initial maps.

Abstract: Apparatuses, systems, and techniques are presented to determine distance and/or motion of one or more objects represented in one or more images. In at least one embodiment, distance and motion are determined using segmentations, single-view depth, and optical flow data inferred for one or more images captured in sequence using a monocular camera.

Abstract: A method of controlling a refrigerator includes instructing at least one camera to capture images forward of a main body opening of a main body of the refrigerator of items being placed into or removed from at least one storage compartment. The images are analyzed to determine whether the images contain an object of interest. The method further includes assigning a direction of movement for the object of interest being at least one of into or out of the storage compartment. An inventory for the storage compartment is updated based on the object of interest and the direction of movement for the object of interest. Instructions are then provided to at least one environmental control device. The instructions control the environmental control device to change a storage condition within the storage compartment based on the inventory in the storage compartment.

Abstract: A computing system and method can be used to render a 3D shape from one or more images. In particular, the present disclosure provides a general pipeline for learning articulated shape reconstruction from images (LASR). The pipeline can reconstruct rigid or nonrigid 3D shapes. In particular, the pipeline can automatically decompose non-rigidly deforming shapes into rigid motions near rigid-bones. This pipeline incorporates an analysis-by-synthesis strategy and forward-renders silhouette, optical flow, and color images which can be compared against the video observations to adjust the internal parameters of the model. By inverting a rendering pipeline and incorporating optical flow, the pipeline can recover a mesh of a 3D model from the one or more images input by a user.

Abstract: Systems and methods to train a convolutional neural network having two or more filter layers having different filtering parameters corresponding to respective different portions of a digital representation of an image. A processor comprising one or more arithmetic logic units (ALUs) to be configured to identify one or more features within an image based, at least in part, on a convolutional neural network having two or more filter layers having different filtering parameters corresponding to respective different portions of a digital representation of the image.

Abstract: A refrigerator system includes a main body and a door mounted to and movable with respect to the main body, an ambient light sensor, a light source, at least one camera positioned to have a field of view that includes an entrance opening leading to the at least one storage compartment, and at least one computing device in operable connection with the at least one camera, the ambient light sensor, and the light source.

Abstract: Despite recent progress, existing frame interpolation methods still struggle with extremely high resolution images and challenging cases such as repetitive textures, thin objects, and fast motion. To address these issues, provided is a cascaded diffusion frame interpolation approach that excels in these scenarios while achieving competitive performance on standard benchmarks.

Abstract: This document describes systems and techniques directed at fusing optically zoomed images into one digitally zoomed image. In aspects, a computing device having at least two cameras and an image-processing manager is configured to receive, from a first camera, a first image at a first optical zoom and, from a second camera, a second image at a second optical zoom different from the first optical zoom. The first and second cameras capture a same scene from different fields of view and different points of view. The image-processing manager receives a desired digital zoom between the first optical zoom and the second optical zoom. Based on the first and second images, the image-processing manager determines an overlap region of the first image in which the second image overlaps the first image.

Abstract: Provided is a video generation model for performing text-to-video (T2V) or other video generation techniques. The proposed model reduces the computational costs associated with video generation. In particular, unlike traditional T2V methods, the disclosed technology can generate the full temporal duration of a video clip at once, bypassing the need for extensive computation. As one example, a machine-learned denoising diffusion model can simultaneously process a plurality of noisy inputs that correspond to various timestamps spanning the temporal dimension of a video to simultaneously generate synthetic frames for the video that match the timestamps.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controllable video generation. One of the methods includes receiving a text prompt that specifies an object; receiving a control input that comprises an image that depicts a particular instance of the object; generating a video that comprises a respective video frame at each of a plurality of time steps in the video and that depicts the particular instance of the object. Generating the video includes, at each of the plurality of time steps: obtaining a text prompt embedding; obtaining a control input embedding; and generating the respective video frame at the time step using a video generation neural network while the video generation neural network is conditioned on the text prompt embedding and on the control input embedding.

Abstract: Provided is a computer system that includes one or more processors and one or more non-transitory computer-readable media that collectively store a machine-learned image interpolation model. The machine-learned image interpolation model is configured to: extract, for each of multiple different scales, a respective set of feature values from each of a pair of input images; generate, for each of the multiple different scales, a respective flow estimate for each of the pair of input images that indicates a respective flow from the interpolation time to the respective capture time; warp, for each of the multiple different scales, the respective set of feature values for each of the pair of input images according to the respective flow estimate to generate respective warped sets of features; and generate a interpolated image based on the respective warped sets of features for the pair of input images and the multiple different scales.

Abstract: Systems and methods for training models to predict dense correspondences across images such as human images. A model may be trained using synthetic training data created from one or more 3D computer models of a subject. In addition, one or more geodesic distances derived from the surfaces of one or more of the 3D models may be used to generate one or more loss values, which may in turn be used in modifying the model's parameters during training.

Abstract: Systems and methods for training models to predict dense correspondences across images such as human images. A model may be trained using synthetic training data created from one or more 3D computer models of a subject. In addition, one or more geodesic distances derived from the surfaces of one or more of the 3D models may be used to generate one or more loss values, which may in turn be used in modifying the model's parameters during training.

Abstract: Systems and methods for training models to predict dense correspondences across images such as human images. A model may be trained using synthetic training data created from one or more 3D computer models of a subject. In addition, one or more geodesic distances derived from the surfaces of one or more of the 3D models may be used to generate one or more loss values, which may in turn be used in modifying the model's parameters during training.

Abstract: A computing system and method can be used to render a 3D shape from one or more images. In particular, the present disclosure provides a general pipeline for learning articulated shape reconstruction from images (LASR). The pipeline can reconstruct rigid or nonrigid 3D shapes. In particular, the pipeline can automatically decompose non-rigidly deforming shapes into rigid motions near rigid-bones. This pipeline incorporates an analysis-by-synthesis strategy and forward-renders silhouette, optical flow, and color images which can be compared against the video observations to adjust the internal parameters of the model. By inverting a rendering pipeline and incorporating optical flow, the pipeline can recover a mesh of a 3D model from the one or more images input by a user.

Abstract: A method includes obtaining a first plurality of feature vectors associated with a first image and a second plurality of feature vectors associated with a second image. The method also includes generating a plurality of transformed feature vectors by transforming each respective feature vector of the first plurality of feature vectors by a kernel matrix trained to define an elliptical inner product space. The method additionally includes generating a cost volume by determining, for each respective transformed feature vector of the plurality of transformed feature vectors, a plurality of inner products, wherein each respective inner product of the plurality of inner products is between the respective transformed feature vector and a corresponding candidate feature vector of a corresponding subset of the second plurality of feature vectors. The method further includes determining, based on the cost volume, a pixel correspondence between the first image and the second image.

Abstract: A method includes filtering a point cloud transformation of a 3D object to generate a 3D lattice and processing the 3D lattice through a series of bilateral convolution networks (BCL), each BCL in the series having a lower lattice feature scale than a preceding BCL in the series. The output of each BCL in the series is concatenated to generate an intermediate 3D lattice. Further filtering of the intermediate 3D lattice generates a first prediction of features of the 3D object.

Abstract: A neural network model receives color data for a sequence of images corresponding to a dynamic scene in three-dimensional (3D) space. Motion of objects in the image sequence results from a combination of a dynamic camera orientation and motion or a change in the shape of an object in the 3D space. The neural network model generates two components that are used to produce a 3D motion field representing the dynamic (non-rigid) part of the scene. The two components are information identifying dynamic and static portions of each image and the camera orientation. The dynamic portions of each image contain motion in the 3D space that is independent of the camera orientation. In other words, the motion in the 3D space (estimated 3D scene flow data) is separated from the motion of the camera.

Abstract: A method, computer readable medium, and system are disclosed for identifying residual video data. This data describes data that is lost during a compression of original video data. For example, the original video data may be compressed and then decompressed, and this result may be compared to the original video data to determine the residual video data. This residual video data is transformed into a smaller format by means of encoding, binarizing, and compressing, and is sent to a destination. At the destination, the residual video data is transformed back into its original format and is used during the decompression of the compressed original video data to improve a quality of the decompressed original video data.

Abstract: A method includes obtaining a first plurality of feature vectors associated with a first image and a second plurality of feature vectors associated with a second image. The method also includes generating a plurality of transformed feature vectors by transforming each respective feature vector of the first plurality of feature vectors by a kernel matrix trained to define an elliptical inner product space. The method additionally includes generating a cost volume by determining, for each respective transformed feature vector of the plurality of transformed feature vectors, a plurality of inner products, wherein each respective inner product of the plurality of inner products is between the respective transformed feature vector and a corresponding candidate feature vector of a corresponding subset of the second plurality of feature vectors. The method further includes determining, based on the cost volume, a pixel correspondence between the first image and the second image.