Abstract: With Convolutional Neural Networks (CNN), facial alignment networks (FAN) have achieved significant accuracy on a wide range of public datasets, which comes along with larger model size and expensive computation costs, making it infeasible to adapt them to real-time applications on edge devices. There is provided a model compression approach on FAN using One-Shot Neural Architecture Search to overcome this problem while preserving performance criteria. Methods and devices provide efficient training and searching (on a single GPU), and resultant models can deploy to run real-time in browser-based applications on edge devices including tablets and smartphones. The compressed models provide comparable cutting-edge accuracy, while having a 30 times smaller model size and can run 40.7 ms per frame in a popular browser on a popular smartphone and OS.

Abstract: Provided in the present disclosure are a video generation method and apparatus, and a device and a storage medium. The method includes: firstly, when an application request for a target effect object has been received, acquiring location information of the current user; then, on the basis of the location information of the current user, determining a target place of interest (POI) corresponding to the current user; acquiring an effect resource of the target POI, and on the basis of the location information of the current user and location information of the target POI, determining a direction relationship between the current user and the target POI; on the basis of the direction relationship, displaying, on a photographic page, a photographic effect corresponding to the effect resource; and then on the basis of the photographic effect corresponding to the effect resource, generating a result video corresponding to the target effect object.

Abstract: Provided are a sound effect display method and a terminal device. The method includes receiving a first operation input for a target audio file; and in response to the first operation, displaying a sound effect display interface corresponding to the target audio file, the sound effect display interface includes a foreground image and a background image, the foreground image includes a cover image corresponding to the target audio file, the background image is an image obtained by photographing a particle dot matrix at a preset photographic angle, the particle dot matrix includes a plurality of particle dots, initial positions of which are located on a first plane, and each particle dot of the particle dot matrix moves, according to a preset sound attribute of the target audio file, in the direction perpendicular to the first plane.

Abstract: A sound effect display method and a terminal device.

Abstract: Provided are a sound effect display method and a terminal device. The method includes receiving a first operation input to a target audio file; in response to the first operation, displaying a sound effect display interface corresponding to the target audio file, the sound effect display interface includes a grid, a sound wave line, and a cover image corresponding to the target audio file; the grid is located on a first plane in an interface scenario of the sound effect display interface, and the first plane is a horizontal plane in the interface scenario; the sound wave line is a closed curve in the first plane, and is used for displaying a sound wave waveform corresponding to a preset sound attribute of the target audio file; and the cover image is on a second plane in the interface scenario.

Abstract: Methods and systems for enhancing elevation models with landcover feature data are provided. An example method involves accessing imagery covering an area of interest that contains an above-ground feature, generating, a digital surface model of the area of interest, extracting three-dimensional landcover feature data representing the above-ground feature that contains one or more actual ground-level points at the base of the above-ground feature, and generating a digital terrain model from the digital surface model at least in part by removing the above-ground feature from the digital surface model and filling at least some of the removed area with one or more of the actual ground-level points from the three-dimensional landcover feature data.

Abstract: Provided are a beveled light-emitting lamp bead, a beveled light-emitting lamp bead holder, a manufacturing method therefor, and a display device. The beveled light-emitting lamp bead holder includes a holder body and a metal holder embedded in the holder body, where the holder body is divided by the metal holder into a cup shield located on a top side and a support structure located on a bottom side, a cup bottom in the cup shield is provided with a chip mounting surface parallel to a cup rim top surface, a bottom portion of the support structure is provided with a soldering plane, the soldering plane is electrically connected to the chip mounting surface through the metal holder, and an inclination angle is formed between the chip mounting surface and the soldering plane.

Abstract: Embodiments of the disclosure provide a method and a video processing apparatus, an electronic device, a computer-readable storage medium, a computer program product, and a computer program. A terminal device obtains coordinate information of a target part of a collected video picture in response to a selection operation for a first effect; displays the first effect on the target part of the video picture according to the coordinate information; transmits a collected video and the coordinate information of the target part in each video picture of the video are transmitted to a server in response to a collection completion instruction for the video. The server adds a second effect corresponding to the first effect to the target part of each video picture according to the coordinate information.

Abstract: Methods and systems for determining the height of structures based on imagery of the structures and associated two-dimensional vector data are provided. An example method involves projecting two-dimensional vector data outlining a roof of a structure into images of the structure captured from different perspectives and feature matching the vector data across the imagery to determine a best-matching three-dimensional position for the roof situated at the height of the structure.

Abstract: The present disclosure relates to an interactive content generation method and apparatus, and a storage medium and an electronic device. The method includes: displaying a content editing interface, wherein the content editing interface includes a canvas area, a material area and an interaction event area; displaying a target interactive scene image corresponding to a scene adding operation in the canvas area in response to the scene adding operation performed by a user in the content editing interface; adding, in response to a selection operation for a target material in the material area performed by the user, the target material to the target interactive scene image, and adding, in response to an interaction event configuration operation for the target material performed by the user in the interaction event area, an interaction event to the target material to obtain target interactive content.

Abstract: Provided are an interaction method and apparatus in a live streaming room, and a device and a storage medium. The method comprises: in response to a trigger operation for a preset duet entry on a live streaming room page, jumping from the live streaming room page to a camera capture page, wherein a host portrait cutout corresponding to the live streaming room page is displayed on the camera capture page and a capture control is provided on the camera capture page acquiring multimedia information captured by a camera, in response to a trigger operation for the capture control on the camera capture page, wherein the multimedia information comprises an image or a video; and synthesizing the multimedia information with the host portrait cutout to obtain a duet multimedia product.

Abstract: The present disclosure describes techniques of implementing augmented reality. The techniques comprise obtaining a target object image in a real-time scene image captured by an image acquisition unit in response to an instruction from a user; generates a virtual image corresponding to a target object based on the target object image and an animation material; and displays the virtual image corresponding to the target object in the real-time scene.

Abstract: Provided are an interaction method and apparatus in a live streaming room, a device, and a storage medium. The method comprises: in response to a trigger operation for a preset drawing entry on a live streaming room page, jumping to a graphic drawing page from the live streaming room page, a drawing trajectory set for a preset object being displayed on the graphical drawing page; when a drawing stroke on the graphic drawing page is received, matching the drawing stroke with the drawing trajectory; and if it is determined that the drawing stroke is successfully matched with the drawing trajectory, displaying prompt information about successful participation in a preset activity, the preset activity and the preset drawing entry having a correspondence.

Abstract: Provided are a beveled light-emitting lamp bead, a beveled light-emitting lamp bead holder, a manufacturing method therefor, and a display device. The beveled light-emitting lamp bead holder includes a holder body and a metal holder embedded in the holder body, where the holder body is divided by the metal holder into a cup shield located on a top side and a support structure located on a bottom side, a cup bottom in the cup shield is provided with a chip mounting surface parallel to a cup rim top surface, a bottom portion of the support structure is provided with a soldering plane, the soldering plane is electrically connected to the chip mounting surface through the metal holder, and an inclination angle is formed between the chip mounting surface and the soldering plane.

Abstract: The present disclosure provides a live-stream interaction method, a live-stream interaction system and related devices. The system includes a terminal of an audience, a terminal of an anchor, and a server; the terminal of the audience is configured to present a live-stream interface, and the live-stream interface includes an interaction control that supports interaction, and when the audience triggers an interactive behavior through the interaction control, an interaction image when the audience performs the interactive behavior is acquired; the terminal of the anchor is configured to send the behavior feature of the anchor to the server; the server is configured to acquire the behavior feature of the audience, match the behavior feature of the audience with the behavior feature of the anchor to obtain a matching result, and send a special effect instruction to the terminal of the audience according to the matching result.

Abstract: Provided are a light source module and a backlight display module. The light source module includes a flexible printed circuit board and a plurality of side-emitting LEDs. The flexible printed circuit board extends along a central axis. The plurality of side-emitting LEDs are sequentially arranged along the central axis and bonded to the flexible printed circuit board, and a light-emitting surface of one side-emitting LED of the plurality of side-emitting LEDs is perpendicular to a bonding surface of the flexible printed circuit board.

Abstract: A list information display method and apparatus, an electronic device, and a storage medium are provided, the method includes: displaying list information in response to an information display operation triggered by a user; obtaining a real scene shooting image; and loading a plurality of types of information related to the list information into the real scene shooting image and display the real scene shooting image. Due to the use of virtual enhanced display technology, while the list information is displayed in the real scene shooting image, the plurality of types of information related to the list information are also displayed. On one hand, the user can obtain more information related to the list information while obtaining the list information, which improves the efficiency of information acquisition; and on the other hand, the user can obtain information in the real scene shooting image, which improves the visual effect and interactive experience.

Abstract: Embodiments of the present disclosure provide a display method based on augmented reality, a device, a storage medium, and a program product, a real-time scene image is acquired, then a head image of a target object is acquired from the real-time scene image if the real-time scene image includes a face image of the target object, where the head image of the target object includes the face image of the target object; a virtual image of the target object is generated according to the head image; and the virtual image of the target object is displayed in the real-time scene image based on an augmented reality display technology.

Abstract: Embodiments of the present disclosure provide a display method based on augmented reality, a device, a storage medium, and a program product, a real-time scene image is acquired, then a head image of a target object is acquired from the real-time scene image if the real-time scene image includes a face image of the target object, where the head image of the target object includes the face image of the target object; a virtual image of the target object is generated according to the head image; and the virtual image of the target object is displayed in the real-time scene image based on an augmented reality display technology.

Abstract: Described are examples for providing augmented reality (AR) effects for a video sharing platform including activating, by a client of a video sharing platform, an AR effect for recording video for storing on the video sharing platform, and storing, by the client and on the video sharing platform, a user-specific context associated with the AR effect in an inventory that persists after the AR effect is deactivated, wherein the user-specific context includes an identifier of a user account associated with the user-specific context.