A method of making a maxillary tray appliance comprises obtaining a first virtual model of an edentulous maxillary gum of a patient, and obtaining a virtual component, where the virtual component comprises a virtual contacting surface. The method further comprises receiving the first virtual model, aligning the first virtual model or the virtual component in 3D space, generating a virtual tray appliance, and fabricating a maxillary tray appliance using the virtual tray appliance.

The disclosure provides a detector, and a detecting system and method for dividing energy regions intelligently. The detecting method may comprise: collecting, by a detector, rays transmitted through a detected object and generating a detection signal according to the rays; wherein each column of pixels of the detector comprises one class-A electrode and a plurality of class-B electrodes, and the class-A electrode and the class-B electrodes are arranged sequentially in a moving direction of the detected object, such that the rays transmitted through the detected object firstly enter into the class-A electrode and then into the class-B electrodes; obtaining image data of the detected object based on the detection signal corresponding to the class-A electrode, and estimating a material component of the detected object based on the image data; adjusting one or more thresholds for dividing the energy regions according to the estimated material component; and determining energy regions to which the detection signal corresponding to the class-B electrodes belongs, according to the adjusted one or more thresholds, and calculating a number of signals in each energy region, so as to obtain the image data of the detected object and determine components of the detected object accurately.

A method and device of providing a replica masturbation device is disclosed. Users upload images of their body parts to a central database, and other users use data sets generated based on these images from the central database to produce replica masturbation devices of the user's body parts.

A method for providing a replica masturbation device is disclosed. Users upload images of their body parts, and other users use data sets generated based on these images to produce replica masturbation devices of the user's body parts.

Systems and methods for multi-modal representation learning are described. One or more embodiments provide a visual representation learning system trained using machine learning techniques. For example, some embodiments of the visual representation learning system are trained using cross-modal training tasks including a combination of intra-modal and inter-modal similarity preservation objectives. In some examples, the training tasks are based on contrastive learning techniques.