14 results about "Visible object" patented technology

The invention discloses a rendering method, which comprises the following steps: constructing an octree, and storing a to-be-rendered object in a to-be-rendered three-dimensional scene in a child node of the octree to obtain a space management octree; dividing the view cone into different view cone levels; traversing the space management octree, and determining a first intersection part between the space of the child node and the view cone and a view cone level to which the first intersection part belongs; distributing the first intersection part to different rendering queues according to the view cone hierarchy to which the first intersection part belongs; shielding and removing the first intersection parts in the different rendering queues to obtain current visible objects to be rendered; and rendering the current visible to-be-rendered object. According to the method, the octree is constructed to manage the three-dimensional scene space, the view cone levels are divided, the rendering queues are distributed, the view cone removing and multi-level shielding removing technologies are combined, the visible objects are accurately judged, the rendering load is effectively reduced, and the method has the advantages that the rendering performance is improved, space management is optimized, and the complex scene processing efficiency is improved.

The invention relates to the technical field of image rendering, and discloses a three-dimensional scene rendering method, device and equipment based on a graphics processor and a medium, the method is applied to the graphics processor, and the method comprises the following steps: receiving instance attribute data sent by a central processing unit; calling a calculation shader to distribute a calculation thread group according to the number of objects in the instance attribute data; calculating a corresponding transformation matrix for each object in parallel through the calculation thread group; adding the instance attribute data and the transformation matrix into an instance buffer area to obtain to-be-rendered instance data of each object; extracting to-be-rendered instance data from the instance buffer area, and performing invisible object removal processing on the to-be-rendered instance data through the calculation thread group to obtain target instance data of a visible object; and rendering the target instance data. According to the invention, the rendering effect and rendering efficiency of the three-dimensional scene are improved.

The application provides a 4D person-object interaction generation method based on a text description, which comprises the following steps: in stage one, 3D person-object interaction key frame recovery: a human body motion sequence is obtained through a human body motion model and uniformly down-sampled to extract a key frame; a human body grid is reconstructed through an SMPL-X model for each key frame to extract a vertex position and form a human body point cloud; an object position anchoring network takes the human body point cloud, an object template point cloud and a text prompt as input to predict an object position and generate a 3D person-object interaction key frame; in stage two, 4D person-object interaction sequence generation: a contact perception diffusion model is constructed to take a sparse 3D person-object interaction key frame as input, extract a conditional signal through a contact perception encoder; based on the conditional signal, the 3D person-object interaction key frame is subjected to time series interpolation through the contact perception diffusion model to generate a time series coherent dense 4D person-object interaction sequence. The application realizes natural and realistic 4D person-object interaction synthesis of an unseen object.

A raster-based method for generating non-planar reflections in real-time. The method is based on projecting each time all visible object's polygons on a single polygon of a non-planar reflecting surface. All the projections are transformed to camera view and moved through the raster pipeline creating a complete, mirrored image.

This invention provides a method for determining what a driver should be looking at while driving. [Solution] In the method for determining the object to be seen, the driver's line of sight is detected (S11), the gaze area is detected based on the detected line of sight of the driver (S12), an image obtained by photographing the area around the vehicle is acquired (S13), a candidate object to be seen is detected by recognizing the object contained in the image (S14), a traffic scene is determined (S15), it is determined whether or not the candidate object to be seen corresponding to the determined traffic scene is included in the gaze area (S16), if the candidate object to be seen corresponding to the determined traffic scene is included in the gaze area, it is determined that the driver performed appropriate visual actions in the determined traffic scene (S17), and if the candidate object to be seen corresponding to the determined traffic scene is not included in the gaze area, it is determined that the driver did not perform appropriate visual actions in the determined traffic scene (S18).

Two-dimensional (2D) tracking data from multiple camera views is processed to determine three-dimensional (3D) object parameters. Image features are extracted from object data associated with 2D tracking across the camera views. The extracted features for objects visible in multiple views are aggregated and spatial representation tokens are generated based on the object data and reference vectors. Instead of computing birds-eye-view (BEV) features, 3D spatial parameters can be determined directly from the aggregated features and spatial representation tokens. Attention operations analyze relationships between different feature sets while frame-to-frame consistency is maintained in the determined 3D parameters. Multi-view feature aggregation and spatial token generation operate directly on 2D tracking data from surround-view cameras, allowing feature processing without view-specific geometric transformations. Direct 2D-to-3D transformation supports object tracking across different camera configurations and viewing conditions.

This invention discloses a hologram calculation and occlusion processing method based on shared object point sampling. Multiple pixels are selected as sampling center points on the hologram plane to form a sampling grid. For each sampling center point, a cone-shaped beam composed of several rays is emitted to perform object point sampling, filtering out visible object points. For each pixel on the holographic plane, K sampling center points are selected based on its positional relationship with the sampling center points. The union of the visible object point sets is calculated to determine the object point set corresponding to the pixel. Each object point in this set is traversed, and its complex amplitude contribution to the target pixel is calculated. All contributions are accumulated to obtain the light field encoded data of the pixel. The hologram calculation method proposed in this invention can guarantee the coherence between pixels, restore the true occlusion relationship of the model, and reduce the number of samples.

The application belongs to the technical field of automatic driving, and provides a positioning and map construction synchronization method, device, medium and vehicle. The positioning and map construction synchronization method comprises the following steps: acquiring a depth image of a surrounding environment of a vehicle; performing instance segmentation on the depth image of the surrounding environment of the vehicle, and simultaneously tracking a pose background area of an image acquisition device to determine each visible object model; tracking each visible object model and a geometric structure to jointly optimize the geometric shape and the pose of each visible object, and integrating overlap rate data of the joint optimization result and the instance segmentation to determine an initialized new object; based on the tracked initialized new object, the geometric shape and the pose of the initialized new object are jointly optimized again, overlap rate data of the re-joint optimization result and the instance segmentation are integrated again, and finally the object with the most accurate geometric shape and pose is integrated, and is positioned and displayed in an instantiated map synchronously.

The objective is to provide a display control device, a display control method, and a display control program that can improve the safety of persons present around invisible objects. The display control device (410) includes an object information acquisition unit (13) that acquires object information, which is information relating to an invisible object (T), and a display control unit (40) that generates an object information image (I1) showing the object information and controls a display device (210) using augmented reality technology so that the object information image (I1) is displayed in a way that is visible to a person (U) present in the vicinity of the invisible object (T).

The invention provides a texture resource management method and device, a storage medium and electronic equipment, and relates to the field of computers. The electronic device determines an invisible object leaving the visual area in response to an update operation of a user on the visual area; a target texture corresponding to the invisible object in the video memory is determined, if no target visible object is left in the visible area, video memory resources occupied by the target texture are recycled, and the target visible object is a display object depending on the target texture in the visible area. In this way, the texture resources which are not quoted by the visual object are recognized and released in time, and therefore the risk of excessive consumption of the video memory is relieved.

The application discloses an information display method and device, a storage medium and electronic equipment, wherein the method comprises the following steps: displaying a map page, wherein the map page is a map page with a positioning information as a center point and a first scale; in response to a query operation, obtaining a query result, determining a second scale of the map page based on a display range corresponding to the query result, performing zoom processing on the map page based on the second scale, and marking a visible object corresponding to the query result in the zoomed map page. In the case that the user performs a query operation, the map page is adaptively zoomed according to the query result, and the visible object corresponding to the query result is marked on the zoomed map page, realizing automatic interaction between the user and the map page, avoiding the case that the user manually adjusts the map page to display the query result, and improving the flexibility of information query in the map page.

Systems and methods are provided for filtering and / or blocking unreliable content that might normally be injected in an Augmented Reality (AR) or Virtual Reality (VR) device. In one implementation, a method includes a step of analyzing a scene obtained by an image capture device associated with an Augmented Reality (AR) system to detect one or more visible objects in the scene. The method also includes using identifying characteristics of the one or more visible objects to obtain content related to the one or more visible objects. Before presenting the content in an augmented content overlay with respect to the AR system, the method further includes filtering the content based on one or more trust or validity factors.

A method for irradiating an image in an optical see-through (OST) head mounted display (HMD) for viewing through, the OST HMD by a user's eye, an object having at least one of known orientation and position and orientation (O / P&O), associated with a first reference frame, the method comprising: generating and irradiating said image for appearing to said user superimposed in an aligned manner to said object, according to predetermined information, eyeball feature position data, and said O / P&O; said predetermined information relates correction data with a plurality of different respective position data values of at least one eyeball feature position of said eye; said predetermined information further includes display corrections of said electro-optical display module with respect to said position data values of said at least one eyeball feature position, with respect to a second reference frame; and said O / P&O is between said second reference and said first reference frame.

The application discloses a learning object attention analysis method, a terminal and a device in an educational meta-universe, generates a visible object set corresponding to a learning object according to a virtual scene of an educational meta-universe space, and determines geometric-attribute information of each visible object in the visible object set; obtains a learning video of the learning object, determines a line-of-sight direction and a micro-expression of the learning object according to the learning video; determines a target visible object that the learning object gazes at according to the geometric-attribute information of each visible object and the line-of-sight direction, and associates the target visible object with the micro-expression; and analyzes the attention of the learning object according to the target visible object and the associated micro-expression, so that the attention of the learning object in the educational meta-universe can be comprehensively and accurately analyzed.