Abstract: Described herein are systems and methods for training machine learning models to generate three-dimensional (3D) motions based on light detection and ranging (LiDAR) point clouds. In various embodiments, a computing system can encode a machine learning model representing an object in a scene. The computing system can train the machine learning model using a dataset comprising synchronous LiDAR point clouds captured by monocular LiDAR sensors and ground-truth three-dimensional motions obtained from IMU devices. The machine learning model can be configured to generate a three-dimensional motion of the object based on an input of a plurality of point cloud frames captured by a monocular LiDAR sensor.

Abstract: Described herein are methods and non-transitory computer-readable media of a computing system configured to obtain a plurality of images of an object from a plurality of orientations at a plurality of times. A machine learning model is encoded to represent a continuous density field of the object that maps a spatial coordinate to a density value. The machine learning model comprises a deformation module configured to deform the spatial coordinate in accordance with a timestamp and a trained deformation weight. The machine learning model further comprises a neural radiance module configured to derive the density value in accordance with the deformed spatial coordinate, the timestamp, a direction, and a trained radiance weight. The machine learning model is trained using the plurality of images. A three-dimensional structure of the object is constructed based on the trained machine learning model.

Abstract: An image rendering system for rendering two-dimensional images in real-time. The image rendering system can receive an implicit representation model of a three-dimensional image. The image rendering system can construct, based on voxel coordinates, a three-dimensional image based on the implicit representation model. The image rendering system can rotate the three-dimensional image to an orientation in a computing space based on a user input. The image rendering system can generate a two-dimensional image based on the rotated three-dimensional image.

Abstract: Embodiments of the present disclosure provide methods and systems for controlling an imaging device, and detecting parameters of a detector. The method may include applying a voltage on a semiconductor crystal of a detector of the imaging device; obtaining a current of the semiconductor crystal; determining, based on the voltage, the current, and at least one preset relationship, at least one parameter of the semiconductor crystal; and calibrating and/or controlling the imaging device based on the at least one parameter.

Abstract: Detector systems are provided. The detector system may include a plurality of edge-on detector modules. Each edge-on detector module may include: a silicon substrate including a front side corresponding to a first side of the each edge-on detector module and a rear side corresponding to a second side of the each edge-on detector module; a plurality of detection elements disposed on the front side of the silicon substrate; a backside electrode disposed on the rear side of the silicon substrate; and/or an anti-scatter structure disposed on at least one of the first side or the second side of the each edge-on detector module, the anti-scatter structure being configured to prevent or reduce scattering of photons emitted into the silicon substrate. The silicon substrate, the plurality of detection elements, the backside electrode, and/or the anti-scatter structure may be configured as an integral piece.

Abstract: A computing system for encoding a machine learning model comprises a plurality of layers and a plurality of computation units. A first set of computation units are configured to process data at a first bit width. A second set of computation units are configured to process at a second bit width. The first bit width is higher than the second bit width. A memory is coupled to the computation units. A controller is coupled to the computation units and the memory. The controller is configured to provide instructions for encoding the machine learning model. The first set of computation units are configured to compute a first set of layers and the second set of computation units are configured to compute a second set of layers.

Abstract: A computer-implemented method of generating editable free-viewport videos is provided. A plurality of video of a scene from a plurality of views is obtained. The scene comprises includes an environment and one or more dynamic entities. A 3D bounding-box is generated for each dynamic entity in the scene. A computer device encodes a machine learning model including an environment layer and a dynamic entity layer for each dynamic entity in the scene. The environment layer represents a continuous function of space and time of the environment. The dynamic entity layer represents a continuous function of space and time of the dynamic entity. The dynamic entity layer includes a deformation module and a neural radiance module. The deformation module is configured to deform a spatial coordinate in accordance with a timestamp and a trained deformation weight. The neural radiance module is configured to derive a density value and a color.

Abstract: A computing core for rendering an image computing core comprises a position encoding logic and a plurality of pipeline logics connected in series in a pipeline. The position encoding logic is configured to transform coordinates and directions of sampling points corresponding to a portion of the image into high dimensional representations. The plurality of pipeline logics are configured to output, based on the high dimensional representation of the coordinates and the high dimensional representation of the directions, intensity and color values of pixels corresponding to the portion of the image in one pipeline cycle. The plurality of pipeline logics are configured to run in parallel.

Abstract: Described herein are methods and non-transitory computer-readable media configured to obtain a plurality of images from a plurality of image scanning orientations for an object. A rigid registration is performed to the plurality of images to obtain a transformation matrix to normalize the plurality of images from their respective image spaces to a normalized image space. Each normalized image comprises a plurality of voxels. A machine learning model comprising an implicit representation of a high-resolution image is trained using the normalized images, wherein the high-resolution image comprises more voxels than the voxels in the normalized images. The high-resolution image is generated based on the trained machine learning model. The plurality of images are a plurality of anisotropic 2D images, while the high resolution image can be a 2D or 3D high resolution image.

Abstract: Provided are a YOLOv5-based real-time detection method and device for blade cracks in aeroengine operation and maintenance. The method includes: sending a first instruction to obtain images of internal blades of an engine; preprocessing the images of the internal blades of the engine to obtain a test dataset, a training dataset, and a validation dataset; inputting the training dataset into a preset YOLOv5 network model for training, preliminarily evaluating a model effect derived from training by using the validation dataset to adjust the model, testing the model by using a trained weight file and the test dataset, and obtaining an mAP value and a precision-recall curve to finally evaluate the model; obtaining the images of the internal blades of the engine in real time, detecting the internal blades of the engine in real time by using the weight file, and outputting a detection result.

Abstract: A novel neural modeling framework Neural Transient Field (NeTF) is provided for non-line-of-sight (NLOS) imaging. NeTF recovers the 5D transient function in both spatial location and direction, and the training data input is parametrized on the spherical wave-fronts. A Markov chain Monte Carlo (MCMC) algorithm is used to account for sparse and unbalanced sampling in NeTF.

Abstract: A deep neural network based hair rendering system is presented to model high frequency component of furry objects. Compared with existing approaches, the present method can generate photo-realistic rendering results. An acceleration method is applied in our framework, which can speed up training and rendering processes. In addition, a patch-based training scheme is introduced, which significantly increases the quality of outputs and preserves high frequency details.

Abstract: A computer-implemented method is provided. The method includes obtaining a plurality of images representing projections of an object placed in a plurality of poses and a plurality of translations; assigning a pose embedding vector, a flow embedding vector and a contrast transfer function (CTF) embedding vector to each image; encoding, by a computer device, a machine learning model comprising a pose network, a flow network, a density network and a CTF network; training the machine learning model using the plurality of images; and reconstructing a 3D structure of the object based on the trained machine learning module.

Abstract: An image rendering system comprising a preprocessing unit coupled to a feature extract unit and a color rendering unit over a data bus. The preprocessing unit generates vector representations of spatial coordinates of sample points along camera rays corresponding to pixels of an image to be rendered. The feature extract unit generates a feature map of the image based on the vector representations, color and intensity values of the sample point through a first machine learning model. The color rendering unit renders the image based on the feature map through a second machine learning model. The first machine learning model is different from the second machine learning model.

Abstract: A method of processing light field images for separating a transmitted layer from a reflection layer. The method comprises capturing a plurality of views at a plurality of viewpoints with different polarization angles; obtaining an initial disparity estimation for a first view using SIFT-flow, and warping the first view to a reference view; optimizing an objective function comprising a transmitted layer and a secondary layer using an Augmented Lagrange Multiplier (ALM) with Alternating Direction Minimizing (ADM) strategy; updating the disparity estimation for the first view; repeating the steps of optimizing the objective function and updating the disparity estimation until the change in the objective function between two consecutive iterations is below a threshold; and separating the transmitted layer and the secondary layer using the disparity estimation for the first view.

Abstract: An image-based method of modeling and rendering a three-dimensional model of an object is provided. The method comprises: obtaining a three-dimensional point cloud at each frame of a synchronized, multi-view video of an object, wherein the video comprises a plurality of frames; extracting a feature descriptor for each point in the point cloud for the plurality of frames without storing the feature descriptor for each frame; producing a two-dimensional feature map for a target camera; and using an anti-aliased convolutional neural network to decode the feature map into an image and a foreground mask.

Abstract: A computer-implemented method includes encoding a radiance field of an object onto a machine learning model; conducting, based on a set of training images of the object, a training process on the machine learning model to obtain a trained machine learning model, wherein the training process includes a first training process using a plurality of first test sample points followed by a second training process using a plurality of second test sample points located within a threshold distance from a surface region of the object; obtaining target view parameters indicating a view direction of the object; obtaining a plurality of rays associated with a target image of the object; obtaining render sample points on the plurality of rays associated with the target image; and rendering, by inputting the render sample points to the trained machine learning model, colors associated with the pixels of the target image.

Abstract: According to some embodiments, an imaging processing method for extracting a plurality of planar surfaces from a depth map includes computing a depth change indication map (DCI) from a depth map in accordance with a smoothness threshold. The imaging processing method further includes recursively extracting a plurality of planar region from the depth map, wherein the size of each planar region is dynamically adjusted according to the DCI. The imaging processing method further includes clustering the extracted planar regions into a plurality of groups in accordance with a distance function; and growing each group to generate pixel-wise segmentation results and inlier points statistics simultaneously.

Abstract: Systems, methods, and non-transitory computer-readable media are configured to obtain a set of content items to train a neural radiance field-based (NeRF-based) machine learning model for object recognition. Depth maps of objects depicted in the set of content items can be determined. A first set of training data comprising reconstructed content items depicting only the objects can be generated based on the depth maps. A second set of training data comprising one or more optimal training paths associated with the set of content items can be generated based on the depth maps. The one or more optimal training paths are generated based at least in part on a dissimilarity matrix associated with the set of content items. The NeRF-based machine learning model can be trained based on the first set of training data and the second set of training data.

Abstract: Described herein are systems and methods of capturing motions of humans in a scene. A plurality of IMU devices and a LiDAR sensor are mounted on a human. IMU data is captured by the IMU devices and LiDAR data is captured by the LiDAR sensor. Motions of the human are estimated based on the IMU data and the LiDAR data. A three-dimensional scene map is built based on the LiDAR data. An optimization is performed to obtain optimized motions of the human and optimized scene map.