Abstract: An example system includes: a landmark detection engine to detect landmark positions of landmarks in images based on facial detection; an optical flow landmark engine to determine the landmark positions in the images based on optical flow of the landmarks between the images; a landmark difference engine to determine, for a landmark in a given image: a distance between a detected landmark position and an optical flow landmark position of the landmark; and a weighted landmark determination engine to determine, for a first and second image, a position for the landmark in the second image based on: a respective detected landmark position and a respective optical flow position of the landmark in the second image; and respective distances, determined with the landmark difference engine, between a first detected landmark position of the landmark in the first image and respective optical flow landmark positions for the first and second images.

Abstract: The present application discloses a milk warmer, which includes: a milk warmer base, the milk warmer base is provided with a first accommodation space, and the first accommodation space is used to accommodate milk to be heated; a heating member for heating the first accommodation space; a temperature measuring member for detecting the temperature of the heating member; a first fixing bracket used to fix the position of the temperature measuring member; wherein the heating member is provided with a connection hole, one end of the temperature measuring member is connected in the connection hole, and the other end of the temperature measuring member is connected with the first fixing bracket. The milk warmer of this disclosure can increase the volume of the heating member while maintaining an original volume of the milk warmer, and the heating speed can be increased accordingly.

Abstract: A saliency map of an image is generated. Saliency regions of the saliency map are identified. The saliency regions are merged into a combined saliency region. Candidate image crops of the image are generated based on the combined saliency region. An image crop of the image is selected from the candidate image crops using a machine learning model.

Abstract: An example system includes a landmark engine to detect a facial landmark in an image of a face. The system includes a comparison engine to determine a difference between the facial landmark in the image and a facial landmark of a neutral face. The system also includes an action engine to determine whether a facial action unit occurred based on whether the difference satisfies a condition.

Abstract: The present application relates to milk powder packaging field. The present application provides a powder dispensing apparatus for dispensing fixed amount of powder, including a powder outputting member, a bottle mouth connecting assembly and a plurality of resetting members, the powder outputting member has a powder outlet, the bottle mouth connecting assembly includes a valve and a bottle mouth connecting member, the bottle mouth connecting member has a sliding groove, each the resetting member is located in the sliding groove, one end of each resetting member is connected to the bottom of the sliding groove, the other end is connected to the powder outputting member, the valve and the powder outlet are detachably connected, a side of the bottle mouth connecting member deviating from the valve is connected to the powder storing bottle. The present application improves the accuracy of fixed amount of powder dispensing.

Abstract: An intelligent electrically adjustable dumbbell includes a dumbbell body. The dumbbell body includes a handle; a first and a second weight, assembly. The handle includes a handle body, a driving device, a first telescopic rod and a second telescopic rod. The first and the second telescopic rods are connected to the first and the second weight assemblies respectively. Each of the first and the second weight assemblies includes a plurality of dumbbell pieces. Each dumbbell piece defines a shaft hole. A fastening assembly is connected between two adjacent dumbbell pieces and configured to prevent the two adjacent dumbbell pieces from relative axial movement. The driving device is configured to drive the first and the second telescopic rods to extend into one or more shaft holes so as to mount one or more dumbbell pieces onto two opposite ends of the handle body to adjust weight of the dumbbell body.

Abstract: An example system includes: a landmark detection engine to detect landmark positions of landmarks in images based on facial detection; an optical flow landmark engine to determine the landmark positions in the images based on optical flow of the landmarks between the images; a landmark difference engine to determine, for a landmark in a given image: a distance between a detected landmark position and an optical flow landmark position of the landmark; and a weighted landmark determination engine to determine, for a first and second image, a position for the landmark in the second image based on: a respective detected landmark position and a respective optical flow position of the landmark in the second image; and respective distances, determined with the landmark difference engine, between a first detected landmark position of the landmark in the first image and respective optical flow landmark positions for the first and second images.

Abstract: An example system includes a landmark engine to detect a facial landmark in an image of a face. The system includes a comparison engine to determine a difference between the facial landmark in the image and a facial landmark of a neutral face. The system also includes an action engine to determine whether a facial action unit occurred based on whether the difference satisfies a condition.

Abstract: A reflective dynamic projector includes a light projection device. The light projection device includes a reflective element with a reflective surface, a plurality of first light sources, and a first driving unit. There are bump textures arranged on the reflective surface which make the reflective surface to be a diffuse reflection surface. The reflective surface surrounds the plurality of first light sources and each of the plurality of first light sources is arranged towards the reflective surface; or an outer surface of the reflective element is the reflective surface, the plurality of first light sources surrounds the reflective element, and each of the plurality of first light sources is arranged towards the reflective surface. The first driving unit is connected with the reflective element or the plurality of first light sources and drives the reflective element or the plurality of first light sources to rotate.

Abstract: Pixel blocks of a user-generated print job raster page are clustered into raster color clusters, and pixel blocks of a corresponding scanned page are clustered into scanned color clusters. For each raster color cluster, if a color difference between the cluster and a corresponding scanned color cluster is greater than a threshold, color fading of the cluster is determined from simulated depletion of each colorant of a printing device, and a likely depleted colorant is determined based on the corresponding scanned color cluster and the determined color fading. A most likely depleted colorant is determined from the likely depleted colorant predicted for each raster color cluster for which the determined color difference is greater than the threshold.

Abstract: Pixel blocks of a user-generated print job raster page are clustered into raster color clusters, and pixel blocks of a corresponding scanned page are clustered into scanned color clusters. For each raster color cluster, if a color difference between the cluster and a corresponding scanned color cluster is greater than a threshold, color fading of the cluster is determined from simulated depletion of each colorant of a printing device, and a likely depleted colorant is determined based on the corresponding scanned color cluster and the determined color fading. A most likely depleted colorant is determined from the likely depleted colorant predicted for each raster color cluster for which the determined color difference is greater than the threshold.