Abstract: A method includes determining, based on a first image, a first depth of a first pixel and, based on a second image, a second depth of a second pixel that corresponds to the first pixel. The method also includes determining a first 3D point based on the first depth and a second 3D point based on the second depth, and determining a scene flow between the first and second images. The method additionally includes determining an induced pixel position based on a post-flow 3D point representing the first 3D point displaced according to the scene flow, determining a flow loss value based on the induced pixel position and a position of the second pixel and a depth loss value based on the post-flow 3D point and the second 3D point, and adjusting the depth model or the scene flow model based on the flow and depth loss values.

Abstract: A computer-implemented method for decomposing videos into multiple layers that can be re-combined with modified relative timings can include obtaining video data including image frames depicting objects. For each of the frames, the method can include generating object maps descriptive of a respective location of at least one object of the objects within the image frame. For each of the frames, the image frame and the object maps can be input into a machine-learned layer renderer model. For each of the frames, the method can include receiving, as output from the model, a background layer illustrative of a background of the video data and one or more object layers associated with respective object maps. The object layers can include image data illustrative of the object and associated trace effects such that the one or more object layers and the background layer can be re-combined with modified relative timings.

Abstract: A method includes determining, based on an image having an initial viewpoint, a depth image, and determining a foreground visibility map including visibility values that are inversely proportional to a depth gradient of the depth image. The method also includes determining, based on the depth image, a background disocclusion mask indicating a likelihood that pixel of the image will be disoccluded by a viewpoint adjustment. The method additionally includes generating, based on the image, the depth image, and the background disocclusion mask, an inpainted image and an inpainted depth image. The method further includes generating, based on the depth image and the inpainted depth image, respectively, a first three-dimensional (3D) representation of the image and a second 3D representation of the inpainted image, and generating a modified image having an adjusted viewpoint by combining the first and second 3D representation based on the foreground visibility map.

Abstract: A method includes determining, based on an image having an initial viewpoint, a depth image, and determining a foreground visibility map including visibility values that are inversely proportional to a depth gradient of the depth image. The method also includes determining, based on the depth image, a background disocclusion mask indicating a likelihood that pixel of the image will be disoccluded by a viewpoint adjustment. The method additionally includes generating, based on the image, the depth image, and the background disocclusion mask, an inpainted image and an inpainted depth image. The method further includes generating, based on the depth image and the inpainted depth image, respectively, a first three-dimensional (3D) representation of the image and a second 3D representation of the inpainted image, and generating a modified image having an adjusted viewpoint by combining the first and second 3D representation based on the foreground visibility map.

Abstract: The present disclosure provides a statistical, articulated 3D human shape modeling pipeline within a fully trainable, modular, deep learning framework. In particular, aspects of the present disclosure are directed to a machine-learned 3D human shape model with at least facial and body shape components that are jointly trained end-to-end on a set of training data. Joint training of the model components (e.g., including both facial, hands, and rest of body components) enables improved consistency of synthesis between the generated face and body shapes.

Abstract: The present disclosure provides systems and methods that perform face reconstruction based on an image of a face. In particular, one example system of the present disclosure combines a machine-learned image recognition model with a face modeler that uses a morphable model of a human's facial appearance. The image recognition model can be a deep learning model that generates an embedding in response to receipt of an image (e.g., an uncontrolled image of a face). The example system can further include a small, lightweight, translation model structurally positioned between the image recognition model and the face modeler. The translation model can be a machine-learned model that is trained to receive the embedding generated by the image recognition model and, in response, output a plurality of facial modeling parameter values usable by the face modeler to generate a model of the face.

Abstract: A computer-implemented method for decomposing videos into multiple layers (212, 213) that can be re-combined with modified relative timings includes obtaining video data including a plurality of image frames (201) depicting one or more objects. For each of the plurality of frames, the computer-implemented method includes generating one or more object maps descriptive of a respective location of at least one object of the one or more objects within the image frame. For each of the plurality of frames, the computer-implemented method includes inputting the image frame and the one or more object maps into a machine-learned layer Tenderer model. (220) For each of the plurality of frames, the computer-implemented method includes receiving, as output from the machine-learned layer Tenderer model, a background layer illustrative of a background of the video data and one or more object layers respectively associated with one of the one or more object maps.

Abstract: A method includes determining, based on an image having an initial viewpoint, a depth image, and determining a foreground visibility map including visibility values that are inversely proportional to a depth gradient of the depth image. The method also includes determining, based on the depth image, a background disocclusion mask indicating a likelihood that pixel of the image will be disoccluded by a viewpoint adjustment. The method additionally includes generating, based on the image, the depth image, and the background disocclusion mask, an inpainted image and an inpainted depth image. The method further includes generating, based on the depth image and the inpainted depth image, respectively, a first three-dimensional (3D) representation of the image and a second 3D representation of the inpainted image, and generating a modified image having an adjusted viewpoint by combining the first and second 3D representation based on the foreground visibility map.

Abstract: A method includes determining, based on a first image, a first depth of a first pixel and, based on a second image, a second depth of a second pixel that corresponds to the first pixel. The method also includes determining a first 3D point based on the first depth and a second 3D point based on the second depth, and determining a scene flow between the first and second images. The method additionally includes determining an induced pixel position based on a post-flow 3D point representing the first 3D point displaced according to the scene flow, determining a flow loss value based on the induced pixel position and a position of the second pixel and a depth loss value based on the post-flow 3D point and the second 3D point, and adjusting the depth model or the scene flow model based on the flow and depth loss values.

Abstract: A computer-implemented method for decomposing videos into multiple layers (212, 213) that can be re-combined with modified relative timings includes obtaining video data including a plurality of image frames (201) depicting one or more objects. For each of the plurality of frames, the computer-implemented method includes generating one or more object maps descriptive of a respective location of at least one object of the one or more objects within the image frame. For each of the plurality of frames, the computer-implemented method includes inputting the image frame and the one or more object maps into a machine-learned layer Tenderer model. (220) For each of the plurality of frames, the computer-implemented method includes receiving, as output from the machine-learned layer Tenderer model, a background layer illustrative of a background of the video data and one or more object layers respectively associated with one of the one or more object maps.

Abstract: The present disclosure provides a statistical, articulated 3D human shape modeling pipeline within a fully trainable, modular, deep learning framework. In particular, aspects of the present disclosure are directed to a machine-learned 3D human shape model with at least facial and body shape components that are jointly trained end-to-end on a set of training data. Joint training of the model components (e.g., including both facial, hands, and rest of body components) enables improved consistency of synthesis between the generated face and body shapes.

Abstract: The present disclosure provides systems and methods that perform face reconstruction based on an image of a face. In particular, one example system of the present disclosure combines a machine-learned image recognition model with a face modeler that uses a morphable model of a human's facial appearance. The image recognition model can be a deep learning model that generates an embedding in response to receipt of an image (e.g., an uncontrolled image of a face). The example system can further include a small, lightweight, translation model structurally positioned between the image recognition model and the face modeler. The translation model can be a machine-learned model that is trained to receive the embedding generated by the image recognition model and, in response, output a plurality of facial modeling parameter values usable by the face modeler to generate a model of the face.

Abstract: The present disclosure provides systems and methods that perform face reconstruction based on an image of a face. In particular, one example system of the present disclosure combines a machine-learned image recognition model with a face modeler that uses a morphable model of a human's facial appearance. The image recognition model can be a deep learning model that generates an embedding in response to receipt of an image (e.g., an uncontrolled image of a face). The example system can further include a small, lightweight, translation model structurally positioned between the image recognition model and the face modeler. The translation model can be a machine-learned model that is trained to receive the embedding generated by the image recognition model and, in response, output a plurality of facial modeling parameter values usable by the face modeler to generate a model of the face.

Abstract: Structural health monitoring (SHM) is essential but can be expensive to perform. In an embodiment, a method includes sensing vibrations at a plurality of locations of a structure by a plurality of time-synchronized sensors. The method further includes determining a first set of dependencies of all sensors of the time-synchronized sensors at a first sample time to any sensors of a second sample time, and determining a second set of dependencies of all sensors of the time-synchronized sensors at the second sample time to any sensors of a third sample time. The second sample time is later than the first sample time, and the third sample time is later than the second sample time. The method then determines whether the structure has changed if the first set of dependencies is different from the second set of dependencies. Therefore, automated SHM can ensure safety at a lower cost to building owners.

Abstract: A method and system of converting stereo video content to multi-view video content combines an Eulerian approach with a Lagrangian approach. The method comprises generating a disparity map for each of the left and right views of a received stereoscopic frame. For each corresponding pair of left and right scanlines of the received stereoscopic frame, the method further comprises decomposing the left and right scanlines into a left sum of wavelets or other basis functions, and a right sum wavelets or other basis functions. The method further comprises establishing an initial disparity correspondence between left wavelets and right wavelets based on the generated disparity maps, and refining the initial disparity between the left wavelet and the right wavelet using a phase difference between the corresponding wavelets. The method further comprises reconstructing at least one novel view based on the left and right wavelets.

Abstract: A method and system of converting stereo video content to multi-view video content combines an Eulerian approach with a Lagrangian approach. The method comprises generating a disparity map for each of the left and right views of a received stereoscopic frame. For each corresponding pair of left and right scanlines of the received stereoscopic frame, the method further comprises decomposing the left and right scanlines into a left sum of wavelets or other basis functions, and a right sum wavelets or other basis functions. The method further comprises establishing an initial disparity correspondence between left wavelets and right wavelets based on the generated disparity maps, and refining the initial disparity between the left wavelet and the right wavelet using a phase difference between the corresponding wavelets. The method further comprises reconstructing at least one novel view based on the left and right wavelets.

Abstract: The present disclosure provides systems and methods that perform face reconstruction based on an image of a face. In particular, one example system of the present disclosure combines a machine-learned image recognition model with a face modeler that uses a morphable model of a human's facial appearance. The image recognition model can be a deep learning model that generates an embedding in response to receipt of an image (e.g., an uncontrolled image of a face). The example system can further include a small, lightweight, translation model structurally positioned between the image recognition model and the face modeler. The translation model can be a machine-learned model that is trained to receive the embedding generated by the image recognition model and, in response, output a plurality of facial modeling parameter values usable by the face modeler to generate a model of the face.

Abstract: The present disclosure provides systems and methods that perform face reconstruction based on an image of a face. In particular, one example system of the present disclosure combines a machine-learned image recognition model with a face modeler that uses a morphable model of a human's facial appearance. The image recognition model can be a deep learning model that generates an embedding in response to receipt of an image (e.g., an uncontrolled image of a face). The example system can further include a small, lightweight, translation model structurally positioned between the image recognition model and the face modeler. The translation model can be a machine-learned model that is trained to receive the embedding generated by the image recognition model and, in response, output a plurality of facial modeling parameter values usable by the face modeler to generate a model of the face.

Abstract: A method and system of converting stereo video content to multi-view video content combines an Eulerian approach with a Lagrangian approach. The method comprises generating a disparity map for each of the left and right views of a received stereoscopic frame. For each corresponding pair of left and right scanlines of the received stereoscopic frame, the method further comprises decomposing the left and right scanlines into a left sum of wavelets or other basis functions, and a right sum wavelets or other basis functions. The method further comprises establishing an initial disparity correspondence between left wavelets and right wavelets based on the generated disparity maps, and refining the initial disparity between the left wavelet and the right wavelet using a phase difference between the corresponding wavelets. The method further comprises reconstructing at least one novel view based on the left and right wavelets.

Abstract: An apparatus according to an embodiment of the present invention enables measurement and visualization of a refractive field such as a fluid. An embodiment device obtains video captured by a video camera with an imaging plane. Representations of apparent motions in the video are correlated to determine actual motions of the refractive field. A textured background of the scene can be modeled as stationary, with a refractive field translating between background and video camera. This approach offers multiple advantages over conventional fluid flow visualization, including an ability to use ordinary video equipment outside a laboratory without particle injection. Even natural backgrounds can be used, and fluid motion can be distinguished from refraction changes. Embodiments can render refractive flow visualizations for augmented reality, wearable devices, and video microscopes.