Abstract: Example solutions for image paragraph captioning use a first vision language model to generate visual information (comprising text) for an image. The visual information may include tags, an initial image caption, and information on objects within the image (e.g., further tags and captions, and object attributes and locations within the image). In some examples, the visual information further includes visual clues. A generative language model generates a plurality of image story caption candidates (e.g., descriptive paragraphs) from the visual information. A second vision language model evaluates the plurality of image story caption candidates and selects a caption as the final output caption.

Abstract: Disclosed are an apparatus and a method of low-latency, low-power eye tracking. In some embodiments, the eye tracking method operates a first sensor having a first level of power consumption that tracks positions of an eye of a user. In response to detection that the eye does not change position for a time period, the method stops operation of the first sensor and instead operates a second sensor that detects a change of position of the eye. The second sensor has a level of power consumption lower than the level of power consumption of the first sensor. Once the eye position changes, the second sensor resumes operation.

Abstract: Disclosed are an apparatus and a method of low-latency, low-power eye tracking. In some embodiments, the eye tracking method operates a first sensor having a first level of power consumption that tracks positions of an eye of a user. In response to detection that the eye does not change position for a time period, the method stops operation of the first sensor and instead operates a second sensor that detects a change of position of the eye. The second sensor has a level of power consumption lower than the level of power consumption of the first sensor. Once the eye position changes, the second sensor resumes operation.