A device and computer implemented method for digital image content recognition. The method includes determining, depending on a digital image, a first candidate class for the content of the digital image by a baseline model neural network comprising a first feature extractor and a first classifier for classifying digital images; determining a second candidate class for the content of the digital image by a prototypical neural network comprising a second feature extractor and a second classifier for classifying digital images, classifying the content of the digital image into either the first candidate class or the second candidate class depending on the result of a comparison of a first confidence score for the first candidate class to a threshold or of a comparison of a first confidence score for the first candidate class to a second confidence score for the second candidate class.

Procedural model digital content editing techniques are described that overcome the limitations of conventional techniques to make procedural models available for interaction by a wide range of users without requiring specialized knowledge and do so without “breaking” the underlying model. In the techniques described herein, an inverse procedural model system receives a user input that specifies an edit to digital content generated by a procedural model. Input parameters from these candidate input parameters are selected by the system which cause the digital content generated by the procedural model to incorporate the edit.

A radio frequency (RF) imaging device comprising a display receives a three-dimensional (3D) image that is a superposition of two or more images having different image types including at least a 3D RF image of a space disposed behind a surface. A plurality of input control devices receive a user input for manipulating the display of the 3D image. Alternatively or additionally, the radio frequency (RF) imaging device may receive a three-dimensional (3D) image that is a weighted combination of a plurality of images including a 3D RF image of a space disposed behind a surface, an infrared (IR) image of the surface, and a visible light image of the surface. A user input may specify changes to the weighted combination. In another embodiment, the RF imaging device may include an output device that produces a physical output indicating a detected type of material of an object in the space.

The present disclosure relates to an image processing method. The method may include: obtaining image data; reconstructing an image based on the image data, the image including one or more first edges; obtaining a model, the model including one or more second edges corresponding to the one or more first edges; matching the model and the image; and adjusting the one or more second edges of the model based on the one or more first edges.

An image parsing method includes obtaining feature information of an initial image, parsing first feature information in the feature information using a first channel to obtain a first prediction result, parsing second feature information in the feature information using a second channel to obtain a second prediction result, where a size of the first feature information meets a first size range, a size of the second feature information meets a second size range, and the first size range is less than the second size range, and outputting the first prediction result and the second prediction result as a parsing result of the initial image.

Disclosed are a motion image compensation method and a device, and a display device. The motion image compensation method comprises: acquiring a first image corresponding to a pre-motion picture and a second image corresponding to a post-motion picture; dividing the first image and the second image respectively into several image blocks; performing computations on any pair of mutually corresponding image blocks in the first image and the second image based on a phase correlation method performing edge detection on the first image and the second image respectively and according to a preset condition, acquiring, from the at least one displacement vector, a displacement vector matched with a transformation between the edge feature patterns within the pair of mutually corresponding image blocks as a displacement vector corresponding to the pair of image blocks during a motion compensation process.

Embodiments of the present invention are directed to beautifying freeform input paths in accordance with paths existing in the drawing (i.e., resolved paths). In some embodiments of the present invention, freeform input paths of a curved format can be modified or replaced to more precisely illustrate a path desired by a user. As such, a user can provide a freeform input path that resembles a path of interest by the user, but is not as precise as desired. Based on existing paths in the electronic drawing, a path suggestion(s) can be generated to rectify, modify, or replace the input path with a more precise path. In some cases, a user can then select a desired path suggestion, and the selected path then replaces the initially provided freeform input path.

Various embodiments set forth systems and techniques for training a landmark model. The techniques include determining, using the landmark model, a first landmark in a set of first landmarks associated with a first image; performing, on the first image, a first perturbation to obtain a second image; determining, using the landmark model, a second landmark in a set of second landmarks associated with the second image; determining, based on a first distance between the first landmark and the second landmark, a first loss function; and updating, based on the first loss function, a first parameter of the landmark model.

Real time feedback during move, resize or rotate actions in an electronic document is provided. During live drag actions, the layout shown during the action may be the layout produced when the object is dropped. Detection of slow performance of an application and aborting a redraw action as needed may be provided. A visual indication may be provided for helping users to see that an action is in progress, and thereby removing a perception that the application is failing to respond properly. Determining if a figure will not land in a current cursor/touch point position in a move action and a visual indication to help the user see where and why the figure may jump to a new location may also be provided. In addition, determining an attachment point for an object on a page comprising a table may also be provided.

In one embodiment, a computing device receives a request from a mobile device of a user for an image. The computing device determines whether the user is associated with a preferred orientation for viewing an image. The preferred orientation is determined based on the user's usage patterns with respect to a particular application of the mobile device, or to the user's interactions with a social-networking system. If the user is associated with a preferred orientation, the image is adjusted to have the preferred orientation and to match display capabilities of the mobile device sending the request. The adjusted image is sent to the mobile device for display.