Abstract: Scene flow represents the three-dimensional (3D) structure and movement of objects in a video sequence in three dimensions from frame-to-frame and is used to track objects and estimate speeds for autonomous driving applications. Scene flow is recovered by a neural network system from a video sequence captured from at least two viewpoints (e.g., cameras), such as a left-eye and right-eye of a viewer. An encoder portion of the system extracts features from frames of the video sequence. The features are input to a first decoder to predict optical flow and a second decoder to predict disparity. The optical flow represents pixel movement in (x,y) and the disparity represents pixel movement in z (depth). When combined, the optical flow and disparity represent the scene flow.

Abstract: Training deep neural networks requires a large amount of labeled training data. Conventionally, labeled training data is generated by gathering real images that are manually labelled which is very time-consuming. Instead of manually labelling a training dataset, domain randomization technique is used generate training data that is automatically labeled. The generated training data may be used to train neural networks for object detection and segmentation (labelling) tasks. In an embodiment, the generated training data includes synthetic input images generated by rendering three-dimensional (3D) objects of interest in a 3D scene. In an embodiment, the generated training data includes synthetic input images generated by rendering 3D objects of interest on a 2D background image. The 3D objects of interest are objects that a neural network is trained to detect and/or label.

Abstract: A temporal propagation network (TPN) system learns the affinity matrix for video image processing tasks. An affinity matrix is a generic matrix that defines the similarity of two points in space. The TPN system includes a guidance neural network model and a temporal propagation module and is trained for a particular computer vision task to propagate visual properties from a key-frame represented by dense data (color), to another frame that is represented by coarse data (grey-scale). The guidance neural network model generates an affinity matrix referred to as a global transformation matrix from task-specific data for the key-frame and the other frame. The temporal propagation module applies the global transformation matrix to the key-frame property data to produce propagated property data (color) for the other frame. For example, the TPN system may be used to colorize several frames of greyscale video using a single manually colorized key-frame.

Abstract: Training deep neural networks requires a large amount of labeled training data. Conventionally, labeled training data is generated by gathering real images that are manually labelled which is very time-consuming. Instead of manually labelling a training dataset, domain randomization technique is used generate training data that is automatically labeled. The generated training data may be used to train neural networks for object detection and segmentation (labelling) tasks. In an embodiment, the generated training data includes synthetic input images generated by rendering three-dimensional (3D) objects of interest in a 3D scene. In an embodiment, the generated training data includes synthetic input images generated by rendering 3D objects of interest on a 2D background image. The 3D objects of interest are objects that a neural network is trained to detect and/or label.

Abstract: Segmentation is the identification of separate objects within an image. An example is identification of a pedestrian passing in front of a car, where the pedestrian is a first object and the car is a second object. Superpixel segmentation is the identification of regions of pixels within an object that have similar properties. An example is identification of pixel regions having a similar color, such as different articles of clothing worn by the pedestrian and different components of the car. A pixel affinity neural network (PAN) model is trained to generate pixel affinity maps for superpixel segmentation. The pixel affinity map defines the similarity of two points in space. In an embodiment, the pixel affinity map indicates a horizontal affinity and vertical affinity for each pixel in the image. The pixel affinity map is processed to identify the superpixels.

Abstract: Systems and methods to detect one or more segments of one or more objects within one or more images based, at least in part, on a neural network trained in an unsupervised manner to infer the one or more segments. Systems and methods to help train one or more neural networks to detect one or more segments of one or more objects within one or more images in an unsupervised manner.

Abstract: A superpixel sampling network utilizes a neural network coupled to a differentiable simple linear iterative clustering component to determine pixel-superpixel associations from a set of pixel features output by the neural network. The superpixel sampling network computes updated superpixel centers and final pixel-superpixel associations over a number of iterations.

Abstract: Various types of image analysis benefit from a multi-stream architecture that allows the analysis to consider shape data. A shape stream can process image data in parallel with a primary stream, where data from layers of a network in the primary stream is provided as input to a network of the shape stream. The shape data can be fused with the primary analysis data to produce more accurate output, such as to produce accurate boundary information when the shape data is used with semantic segmentation data produced by the primary stream. A gate structure can be used to connect the intermediate layers of the primary and shape streams, using higher level activations to gate lower level activations in the shape stream. Such a gate structure can help focus the shape stream on the relevant information and reduces any additional weight of the shape stream.

Abstract: Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames. A first neural network model approximates optical flow data defining motion between the two consecutive frames. A second neural network model refines the optical flow data and predicts visibility maps for each timestep. The two consecutive frames are warped according to the refined optical flow data for each timestep to produce pairs of warped frames for each timestep. The second neural network model then fuses the pair of warped frames based on the visibility maps to produce the intermediate frame for each timestep. Artifacts caused by motion boundaries and occlusions are reduced in the predicted intermediate frames.

Abstract: Segmentation is the identification of separate objects within an image. An example is identification of a pedestrian passing in front of a car, where the pedestrian is a first object and the car is a second object. Superpixel segmentation is the identification of regions of pixels within an object that have similar properties An example is identification of pixel regions having a similar color, such as different articles of clothing worn by the pedestrian and different components of the car. A pixel affinity neural network (PAN) model is trained to generate pixel affinity maps for superpixel segmentation. The pixel affinity map defines the similarity of two points in space. In an embodiment, the pixel affinity map indicates a horizontal affinity and vertical affinity for each pixel in the image. The pixel affinity map is processed to identify the superpixels.

Abstract: Scene flow represents the three-dimensional (3D) structure and movement of objects in a video sequence in three dimensions from frame-to-frame and is used to track objects and estimate speeds for autonomous driving applications. Scene flow is recovered by a neural network system from a video sequence captured from at least two viewpoints (e.g., cameras), such as a left-eye and right-eye of a viewer. An encoder portion of the system extracts features from frames of the video sequence. The features are input to a first decoder to predict optical flow and a second decoder to predict disparity. The optical flow represents pixel movement in (x,y) and the disparity represents pixel movement in z (depth). When combined, the optical flow and disparity represent the scene flow.

Abstract: A superpixel sampling network utilizes a neural network coupled to a differentiable simple linear iterative clustering component to determine pixel-superpixel associations from a set of pixel features output by the neural network. The superpixel sampling network computes updated superpixel centers and final pixel-superpixel associations over a number of iterations.

Abstract: Training deep neural networks requires a large amount of labeled training data. Conventionally, labeled training data is generated by gathering real images that are manually labelled which is very time-consuming. Instead of manually labelling a training dataset, domain randomization technique is used generate training data that is automatically labeled. The generated training data may be used to train neural networks for object detection and segmentation (labelling) tasks. In an embodiment, the generated training data includes synthetic input images generated by rendering three-dimensional (3D) objects of interest in a 3D scene. In an embodiment, the generated training data includes synthetic input images generated by rendering 3D objects of interest on a 2D background image. The 3D objects of interest are objects that a neural network is trained to detect and/or label.

Abstract: A temporal propagation network (TPN) system learns the affinity matrix for video image processing tasks. An affinity matrix is a generic matrix that defines the similarity of two points in space. The TPN system includes a guidance neural network model and a temporal propagation module and is trained for a particular computer vision task to propagate visual properties from a key-frame represented by dense data (color), to another frame that is represented by coarse data (grey-scale). The guidance neural network model generates an affinity matrix referred to as a global transformation matrix from task-specific data for the key-frame and the other frame. The temporal propagation module applies the global transformation matrix to the key-frame property data to produce propagated property data (color) for the other frame. For example, the TPN system may be used to colorize several frames of greyscale video using a single manually colorized key-frame.

Abstract: Segmentation is the identification of separate objects within an image. An example is identification of a pedestrian passing in front of a car, where the pedestrian is a first object and the car is a second object. Superpixel segmentation is the identification of regions of pixels within an object that have similar properties An example is identification of pixel regions having a similar color, such as different articles of clothing worn by the pedestrian and different components of the car. A pixel affinity neural network (PAN) model is trained to generate pixel affinity maps for superpixel segmentation. The pixel affinity map defines the similarity of two points in space. In an embodiment, the pixel affinity map indicates a horizonal affinity and vertical affinity for each pixel in the image. The pixel affinity map is processed to identify the superpixels.

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 concatentated to generate an intermediate 3D lattice. Further filtering of the intermediate 3D lattice generates a first prediction of features of the 3D object.

Abstract: Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames. A first neural network model approximates optical flow data defining motion between the two consecutive frames. A second neural network model refines the optical flow data and predicts visibility maps for each timestep. The two consecutive frames are warped according to the refined optical flow data for each timestep to produce pairs of warped frames for each timestep. The second neural network model then fuses the pair of warped frames based on the visibility maps to produce the intermediate frame for each timestep. Artifacts caused by motion boundaries and occlusions are reduced in the predicted intermediate frames.