Abstract: The subject technology provides a framework for generating physiological predictions for a user of an electronic device. The physiological predictions may include user-specific predictions of a heartrate, a heartrate range, a number of steps, a number of calories, or other physiological conditions or aspects that may occur if the user engages in a future activity, such as a future workout. The physiological predictions may be generated by a machine learning model that incorporates a physiological state equation, and that generates, and utilizes, a user-specific embedding, along with user-agnostic parameters of the future activity, to make the predictions.

Abstract: Methods, apparatus, and computer-readable storage media for dynamic color flow modeling. A dynamic color flow model that incorporates motion estimation in a probabilistic fashion is described. By automatically and adaptively changing model parameters based on the inferred local motion uncertainty, the dynamic color flow model accurately and reliably models the object appearance, and improves the foreground color probability estimation when compared to conventional methods. The dynamic color flow model may, for example, be applied to both foreground and background layers (or to additional layers) for video object segmentation, obtaining significantly improved results when compared to conventional methods.

Abstract: Methods, apparatus, and computer-readable storage media for dynamic color flow modeling. A dynamic color flow model that incorporates motion estimation in a probabilistic fashion is described. By automatically and adaptively changing model parameters based on the inferred local motion uncertainty, the dynamic color flow model accurately and reliably models the object appearance, and improves the foreground color probability estimation when compared to conventional methods. The dynamic color flow model may, for example, be applied to both foreground and background layers (or to additional layers) for video object segmentation, obtaining significantly improved results when compared to conventional methods.