SoftEye, Inc.

Novel applications for anamorphic lenses in smart glasses. Anamorphic lenses preserve straightness of motion and lines (i.e., “linearity”). As a practical matter, existing computer vision models can be used with anamorphic images without re-training or intermediate conversion steps (unlike fisheye lenses). The contents of the present disclosure provide substantial improvements for applications that have different FOV requirements along different axis. Solutions for ergonomic hand placement (relative to gaze), non-square photosites, binning and eye-tracking are discussed throughout.

Systems, computer programs, devices, and methods that enable coordination across multiple devices of the mobile ecosystem. In one embodiment, smart glasses detect when a user is about to eat food or take a drink and capture the consumable and portion. The data is recorded in a “morsel track” for health activity analysis. Low-fidelity captures provide preliminary recognition, while higher-fidelity captures are selectively invoked for definitive classification. Machine-learning logic generates predicted metabolic responses, such as real-time glucose trends, based on the recorded events. Predicted responses may dynamically adjust the operation of continuous glucose monitors, heart-rate sensors, or other biomedical devices. In some embodiments, the system triggers a pharmaceutical dispenser, such as an insulin pump, inhaler, or transdermal patch, to provide closed-loop therapeutic intervention in real time.

Methods and apparatus for scalable processing. Conventional image sensors read image data in a sequential row-by-row manner. However, image data may be more efficiently processed at different scales. For example, computer vision processing at a first scale may be used to determine whether subsequent processing with more resolution is helpful. Various embodiments of the present disclosure readout image data according to different scales; scaled readouts may be processed using scale specific computer vision algorithms to determine next steps. In addition to scaled readouts of image data, some variants may also provide commonly used data and/or implement pre-processing steps.

Systems, apparatus, and methods for augmenting vision with region-of-interest based processing. In one specific example, smart glasses may use an eye-tracking camera to monitor the user's gaze and determine the user's gaze point. When triggered, the camera assembly captures a high-resolution image. The high-resolution image may be cropped to a much smaller region-of-interest (ROI) image based on computer-vision analysis of the user's gaze point. For example, if the smart glasses detect a human face at the gaze point, then the ROI is cropped to the human face. In this manner, the smart glasses may leverage specific capabilities of the smart glasses to augment the user experience; for example, telephoto lenses provide long distance vision, or computer-assisted search may direct the user to interesting activity. Other aspects may include e.g., external database assisted operation and/or ongoing cataloging throughout the day.

Systems, apparatus, and methods for a gesture-based augmented reality and/or extended reality (AR/XR) user interface. Conventional image processing scales quadratically based on image resolution. Processing complexity directly corresponds to memory size, power consumption, and heat dissipation. As a result, existing smart glasses solutions have short run-times (<1 hr) and may have battery weight and heat dissipation issues that are uncomfortable for continuous wear. The disclosed solution provides a system and method for low-power image processing via the use of scalable processing. In one specific implementation, gesture detection is divided into multiple stages. Each stage conditionally enables subsequent stages for more complex processing. By scaling processing complexity at each stage, high complexity processing can be performed on an “as-needed” basis.

Methods and apparatus for scalable processing. Conventional image sensors read image data in a sequential row-by-row manner. However, image data may be more efficiently processed at different scales. For example, computer vision processing at a first scale may be used to determine whether subsequent processing with more resolution is helpful. Various embodiments of the present disclosure readout image data according to different scales; scaled readouts may be processed using scale specific computer vision algorithms to determine next steps. In addition to scaled readouts of image data, some variants may also provide commonly used data and/or implement pre-processing steps.

Systems, apparatus, and methods for a gesture-based augmented reality and/or extended reality (AR/XR) user interface. Conventional image processing scales quadratically based on image resolution. Processing complexity directly corresponds to memory size, power consumption, and heat dissipation. As a result, existing smart glasses solutions have short run-times (<1 hr) and may have battery weight and heat dissipation issues that are uncomfortable for continuous wear. The disclosed solution provides a system and method for low-power image processing via the use of scalable processing. In one specific implementation, gesture detection is divided into multiple stages. Each stage conditionally enables subsequent stages for more complex processing. By scaling processing complexity at each stage, high complexity processing can be performed on an “as-needed” basis.

Systems, computer programs, devices, and methods that enable LLM-based user interfaces within real-time and/or embedded devices. Providing user-specific context to a generically trained LLM may enable a variety of new usages and scenarios. For example, adaptive prompt augmentation may enable a user device to augment user-generated prompts with additional user context in the form of machine-generated prompts. In some variants, machine-generated prompts may be further refined to accommodate e.g., foundation model constraints, etc. APIs for user-specific data structures can be used to e.g., optimize for habitual behaviors, user idiosyncrasies, etc. Agentic query construction may enable a user device to operate with autonomy and decision-making capabilities, beyond prompt-response interactions. Stitching (or dreaming) may be used to identify pattern-based associations within high dimensional space (embedding vectors).

Systems, computer programs, devices, and methods that enable LLM-based user interfaces within real-time and/or embedded devices. Providing user-specific context to a generically trained LLM may enable a variety of new usages and scenarios. For example, adaptive prompt augmentation may enable a user device to augment user-generated prompts with additional user context in the form of machine-generated prompts. In some variants, machine-generated prompts may be further refined to accommodate e.g., foundation model constraints, etc. APIs for user-specific data structures can be used to e.g., optimize for habitual behaviors, user idiosyncrasies, etc. Agentic query construction may enable a user device to operate with autonomy and decision-making capabilities, beyond prompt-response interactions. Stitching (or dreaming) may be used to identify pattern-based associations within high dimensional space (embedding vectors).

Systems, computer programs, devices, and methods that enable LLM-based user interfaces within real-time and/or embedded devices. Providing user-specific context to a generically trained LLM may enable a variety of new usages and scenarios. For example, adaptive prompt augmentation may enable a user device to augment user-generated prompts with additional user context in the form of machine-generated prompts. In some variants, machine-generated prompts may be further refined to accommodate e.g., foundation model constraints, etc. APIs for user-specific data structures can be used to e.g., optimize for habitual behaviors, user idiosyncrasies, etc. Agentic query construction may enable a user device to operate with autonomy and decision-making capabilities, beyond prompt-response interactions. Stitching (or dreaming) may be used to identify pattern-based associations within high dimensional space (embedding vectors).