Abstract: Embodiments described herein provide a system for generating semantically accurate synthetic images. During operation, the system generates a first synthetic image using a first artificial intelligence (AI) model and presents the first synthetic image in a user interface. The user interface allows a user to identify image units of the first synthetic image that are semantically irregular. The system then obtains semantic information for the semantically irregular image units from the user via the user interface and generates a second synthetic image using a second AI model based on the semantic information. The second synthetic image can be an improved image compared to the first synthetic image.

Abstract: A method of labeling a dataset includes inputting a testing set comprising a plurality of input data samples into a plurality of pre-trained machine learning models to generate a set of embeddings output by the plurality of pre-trained machine learning models. The method further includes performing an iterative cluster labeling algorithm that includes generating a plurality of clusterings from the set of embeddings, analyzing the plurality of clusterings to identify a target embedding with a highest duster quality, analyzing the target embedding to determine a compactness for each of the plurality of clusterings of the target embedding, and identifying a target cluster among the plurality of clusterings of the target embedding based on the compactness. The method further includes assigning pseudo-labels to the subset of the plurality of input data samples that are members of the target duster.

Abstract: One embodiment provides a system that facilitates efficient collection of training data. During operation, the system obtains, by a recording device, a first image of a physical object in a scene which is associated with a three-dimensional (3D) world coordinate frame. The system marks, on the first image, a plurality of vertices associated with the physical object, wherein a vertex has 3D coordinates based on the 3D world coordinate frame. The system obtains a plurality of second images of the physical object in the scene while changing one or more characteristics of the scene. The system projects the marked vertices on to a respective second image to indicate a two-dimensional (2D) bounding area associated with the physical object.

Abstract: A three-dimensional (3D) printer includes a nozzle and a camera configured to capture a real image or a real video of a liquid metal while the liquid metal is positioned at least partially within the nozzle. The 3D printer also includes a computing system configured to perform operations. The operations include generating a model of the liquid metal positioned at least partially within the nozzle. The operations also include generating a simulated image or a simulated video of the liquid metal positioned at least partially within the nozzle based at least partially upon the model. The operations also include generating a labeled dataset that comprises the simulated image or the simulated video and a first set of parameters. The operations also include reconstructing the liquid metal in the real image or the real video based at least partially upon the labeled dataset.

Abstract: A system determines an input video and a first annotated image from the input video which identifies an object of interest. The system initiates a tracker based on the first annotated image and the input video. The tracker generates, based on the first annotated image and the input video, information including: a sliding window for false positives; a first set of unlabeled images from the input video; and at least two images with corresponding labeled states. A semi-supervised classifier classifies, based on the information, the first set of unlabeled images from the input video. If a first unlabeled image is classified as a false positive, the system reinitiates the tracker based on a second annotated image occurring in a frame prior to a frame with the false positive. The system generates an output video comprising the input video displayed with tracking on the object of interest.

Abstract: A method and system for implementing a task scheduler are provided in a resource constrained computation system that uses meta data provided for each task (e.g. data analysis algorithm or sensor sampling protocol) to determine which tasks should be run in a particular wake cycle, the order in which the tasks are run, and how the tasks are distributed across the available compute resources. When a task successfully completes, it's time of execution is logged in order to provide a reference for when that task should be run again. Task meta data is formatted in a manner to allow for simple integration of new tasks into the processing architecture.

Abstract: A system and method provide a combination of a modular message structure, a priority-based message packing scheme, and a data packet queue management system to optimize the information content of a transmitted message in, for example, the Ocean of Things (OoT) environment. The modular message structure starts with a header that provides critical information and reference points for time and location. The rest of the message is composed of modular data packets, each of which has a data ID section that the message decoder uses for reference when reconstructing the message contents, an optional size section that specifies the length of the following data section if it can contain data of variable length, and a data section that can be compressed in a manner unique to that data type.

Abstract: A method of image annotation includes obtaining a candidate annotation map for an annotation task for an image from each of a set of annotation models wherein each of the candidate annotation maps includes suggested annotations for the image, receiving user selections or modifications of at least one of the suggested annotations from one or more of the candidate annotation maps, and generating a final annotation map based on the user selections or modifications from the one or more of the candidate annotation maps.

Abstract: The present specification relates to image capture. More specifically, it relates to selective image capture for sensor carrying devices or floats deployed, for example, on the open sea. In one form, data is generated on the sensor carrying devices or floats by an on-board Inertial Measurement Unit (IMU) and is used to automatically predict the wave motion of the sea. These predictions are then used to determine an acceptable set of motion parameters that are used to trigger the on-board camera(s). The camera(s) then capture images. One consideration is that images captured at or near the peak of a wave crest with minimal pitch and roll will contain fewer obstructions (such as other waves). Such images provide a view further into the horizon to, for example, monitor maritime sea traffic and other phenomenon. Therefore, the likelihood of capturing interesting objects such as ships, boats, garbage, birds, . . . etc. is increased.

Abstract: A system is provided which obtains images of a physical object captured by an AR recording device in a 3D scene. The system measures a level of diversity of the obtained images, for a respective image, based on at least: a distance and angle; a lighting condition; and a percentage of occlusion. The system generates, based on the level of diversity, a first visualization of additional images to be captured by projecting, on a display of the recording device, first instructions for capturing the additional images using the AR recording device. The system trains a model based on the collected data. The system performs an error analysis on the collected data to estimate an error rate for each image of the collected data. The system generates, based on the error analysis, a second visualization of further images to be captured. The model is further trained based on the collected data.

Abstract: A system is provided which mixes static scene and live annotations for labeled dataset collection. A first recording device obtains a 3D mesh of a scene with physical objects. The first recording device marks, while in a first mode, first annotations for a physical object displayed in the 3D mesh. The system switches to a second mode. The system displays, on the first recording device while in the second mode, the 3D mesh including a first projection indicating a 2D bounding area corresponding to the marked first annotations. The first recording device marks, while in the second mode, second annotations for the physical object or another physical object displayed in the 3D mesh. The system switches to the first mode. The first recording device displays, while in the first mode, the 3D mesh including a second projection indicating a 2D bounding area corresponding to the marked second annotations.

Abstract: Disclosed are methods and systems of controlling the placement of micro-objects on the surface of a micro-assembler. Control patterns may be used to cause phototransistors or electrodes of the micro-assembler to generate dielectrophoretic (DEP) and electrophoretic (EP) forces which may be used to manipulate, move, position, or orient one or more micro-objects on the surface of the micro-assembler. A set of micro-object may be analyzed. Geometric properties of the set of micro-objects may be identified. The set of micro-objects may be divided into multiple sub-sets of micro-objects based on the one or more geometric properties and one or more control patterns.

Abstract: A method of labeling data and training a model is provided. The method includes obtaining a set of images. The set of images includes a first subset and a second subset. The first subset is associated with a first set of labels. The method also includes generating a set of pseudo labels for the set of images and a second set of labels for the second subset based on the first subset, the second subset, a first machine learning model, and a domain adaption model. The method further includes generating second machine learning model. The second machine learning model is generated based on the set of images, the set of pseudo labels, the first set of labels, and the second set of labels. The second set of labels is updated based on one or more inferences generated by the second machine learning model.

Abstract: A system is provided which obtains images of a physical object captured by an AR recording device in a 3D scene. The system measures a level of diversity of the obtained images, for a respective image, based on at least: a distance and angle; a lighting condition; and a percentage of occlusion. The system generates, based on the level of diversity, a first visualization of additional images to be captured by projecting, on a display of the recording device, first instructions for capturing the additional images using the AR recording device. The system trains a model based on the collected data. The system performs an error analysis on the collected data to estimate an error rate for each image of the collected data. The system generates, based on the error analysis, a second visualization of further images to be captured. The model is further trained based on the collected data.

Abstract: A method of image annotation includes selecting a plurality of annotation models related to an annotation task for an image, obtaining a candidate annotation map for the image from each of the plurality of annotation models, and selecting at least one of the candidate annotation maps to be displayed via a user interface, the candidate annotation maps comprising suggested annotations for the image. The method further includes receiving user selections or modifications of at least one of the suggested annotations from the candidate annotation map and generating a final annotation map based on the user selections or modifications.

Abstract: A method of labeling a dataset includes inputting a testing set comprising a plurality of input data samples into a plurality of pre-trained machine learning models to generate a set of embeddings output by the plurality of pre-trained machine learning models. The method further includes performing an iterative cluster labeling algorithm that includes generating a plurality of clusterings from the set of embeddings, analyzing the plurality of clusterings to identify a target embedding with a highest duster quality, analyzing the target embedding to determine a compactness for each of the plurality of clusterings of the target embedding, and identifying a target cluster among the plurality of clusterings of the target embedding based on the compactness. The method further includes assigning pseudo-labels to the subset of the plurality of input data samples that are members of the target duster.

Abstract: Disclosed are methods and systems of controlling the placement of micro-objects on the surface of a micro-assembler. Control patterns may be used to cause electrodes of the micro-assembler to generate dielectrophoretic (DEP) and electrophoretic (EP) forces which may be used to manipulate, move, position, or orient one or more micro-objects on the surface of the micro-assembler. The control patterns may be part of a library of control patterns.

Abstract: One embodiment facilitates generating synthetic data objects using a semi-supervised GAN. During operation, a generator module synthesizes a data object derived from a noise vector and an attribute label. The system passes, to an unsupervised discriminator module, the data object and a set of training objects which are obtained from a training data set. The unsupervised discriminator module calculates: a value indicating a probability that the data object is real; and a latent feature representation of the data object. The system passes the latent feature representation and the attribute label to a supervised discriminator module. The supervised discriminator module calculates a value indicating a probability that the attribute label given the data object is real. The system performs the aforementioned steps iteratively until the generator module produces data objects with a given attribute label which the unsupervised and supervised discriminator modules can no longer identify as fake.

Abstract: A method is provided. The method includes obtaining an enhanced state graph. The enhanced state graph represents a set of objects within an environment and a set of positions of the set of objects. The enhanced state graph includes a set of object nodes, a set of property nodes and a set of goal nodes to represent a set of objectives. The method also includes generating a set of instructions for a set of mechanical systems based on the enhanced state graph. The set of mechanical systems is configured to interact with one or more of the set of objects within the environment. The method further includes operating the set of mechanical systems to achieve the set of objectives based on the set of instructions.

Abstract: A method of labeling a dataset of input samples for a machine learning task includes selecting a plurality of pre-trained machine learning models that are related to a machine learning task. The method further includes processing a plurality of input data samples through each of the pre-trained models to generate a set of embeddings. The method further includes generating a plurality of clusterings from the set of embeddings. The method further includes analyzing, by a processing device, the plurality of clusterings to extract superclusters. The method further includes assigning pseudo-labels to the input samples based on analysis.