Abstract: A method includes extracting, using a backbone of a machine learning model, a plurality of features from an image of a foot and predicting, using a first portion of the machine learning model and based on one or more features of the plurality of features, a first aspect of the foot. The method also includes predicting, using a second portion of the machine learning model and based on one or more features of the plurality of features, a second aspect of the foot different from the first aspect, generating, using at least the first aspect and the second aspect, a two-dimensional model of a shoe, and superimposing the two-dimensional model of the shoe onto the image of the foot.

Abstract: A vaporizer includes: a liquid storage cavity; a porous body, being in fluid communication with the liquid storage cavity to absorb the liquid substrate, and having a vaporization surface; and a resistance heating trajectory, formed on the vaporization surface and configured to heat at least a part of the liquid substrate of the porous body to generate an aerosol, where the vaporization surface is a flat plane, and includes a length direction and a width direction perpendicular to the length direction; the resistance heating trajectory includes a first end and a second end, and extends between the first end and the second end in a zigzag way in the length direction of the vaporization surface; and a distance that the first end and the second end span in the vaporization surface in the length direction is greater than 75% of a length dimension of the vaporization surface.

Abstract: A method for generating an high-dynamic-range (HDR) color image from a dual-exposure-time single-shot HDR color image sensor includes obtaining pixel values generated by a local region of sensor pixels of the image sensor, determining a motion parameter for the local region from pixel values associated with a first color, and demosaicing the pixel values of the local region to determine, for each of three colors, an output value of the images pixel, wherein relative contributions of short-exposure-time pixels and long-exposure-time pixels to the output value are weighted according to the motion parameter.

Abstract: According to an embodiment, a method includes estimating, based at least in part on a first location of a bounding box for a foot in a first frame of a sequence of frames and a second location of the bounding box in a second frame of the sequence of frames, a smoothing transformation that characterizes a movement of the foot in the sequence of frames and smoothing, based at least in part on the smoothing transformation, a movement of a structure of the foot from a first location of the structure in the first frame to a second location of the structure in the second frame. The method also includes superimposing a model of a shoe onto the foot in the sequence of frames based at least in part on the smoothed movement.

Abstract: The mesh tracking described herein involves mesh tracking on 3D face models. In contrast to existing mesh tracking algorithms which generally require user intervention and manipulation, the mesh tracking algorithm is fully automatic once a template mesh is provided. In addition, an eye and mouth boundary detection algorithm is able to better reconstruct the shape of eyes and mouths.

Abstract: The mesh tracking described herein involves mesh tracking on 3D face models. In contrast to existing mesh tracking algorithms which generally require user intervention and manipulation, the mesh tracking algorithm is fully automatic once a template mesh is provided. In addition, an eye and mouth boundary detection algorithm is able to better reconstruct the shape of eyes and mouths.

Abstract: A method for generating an high-dynamic-range (HDR) color image from a dual-exposure-time single-shot HDR color image sensor includes obtaining pixel values generated by a local region of sensor pixels of the image sensor, determining a motion parameter for the local region from pixel values associated with a first color, and demosaicing the pixel values of the local region to determine, for each of three colors, an output value of the images pixel, wherein relative contributions of short-exposure-time pixels and long-exposure-time pixels to the output value are weighted according to the motion parameter.

Abstract: Methods, systems, and devices are disclosed for implementing and fabricating electrochemical solar cells including dye-sensitized and perovskite-sensitized solar cells. In one aspect, a dye-sensitized solar cell device includes a cathode including a metal mesh structure that is optically transmissive and electrically conductive, an anode including a metal base layer that is optically opaque and electrically conductive, one or more layers of a semiconductive oxide coupled to the anode, the one or more layers of the semiconductive oxide including nanostructures having a photosensitive dye material coating, in which the anode generates photoelectric energy based on absorption of light by the photosensitive dye material, and an electrolyte of a substantially transparent substance and formed between the cathode and the one or more layers of a semiconductive oxide.