110 and render game content based at least in part on game state information received from the game system(s) 110. Additionally, the game client device(s) 130 may send indications of player input to the game system(s) 110. Game state information and player input information may be shared between the game client device(s) 130 and the game system(s) 110 using any suitable mechanism, such as application program interfaces (APIs).

The game system(s) 110 may receive inputs from various player(s) 132 and update the state of the online game based thereon. As the state of the online game is updated, the state may be sent to the game client device(s) 130 for rendering online game content to player(s) 132. In this way, the game system(s) 110 may host the online game.

In some cases, the techniques described herein for detecting scene features and mapping scene features to recommended actions can be performed by the game system(s) 110. For example, the game system(s) 110 may be configured to generate a scanning query to determine a geometric feature in a virtual scene, determine a scene feature based on the geometric feature, map the scene feature to a recommended action, generate display data of the avatar performing the recommended action, and provide the display data to the game client device(s) 130. The game client device(s) 130 may then be configured to display the display data to the player.

In some cases, the techniques described herein for detecting scene features and mapping scene features to recommended actions can be performed by the game client device(s) 130. For example, a game client device may be configured to receive data describing a virtual scene from the game system(s) 110. Afterward, the game client device(s) 130 may be configured to generate a scanning query to determine a geometric feature in the received virtual scene, determine a scene feature based on the geometric feature, map the scene feature to a recommended action, and display the avatar performing the recommended action to the player.

FIG. 2 is a flowchart diagram of an example process 200 for controlling a player avatar based on received scene data for a virtual scene. As depicted in FIG. 2, at operation 202, the process 200 includes receiving scene data for a virtual scene.

The virtual scene may be a digital environment created using computer graphics operations. The virtual scene may be rendered in real-time and displayed on a screen or other output device to give the player a visual representation of the game world. The virtual scene may include various elements such as terrain, objects, characters, and other interactive elements with which the player can interact. The virtual scene may also include lighting, sound effects, and other immersive features that enhance the player's experience.

In some cases, the virtual scene is a simulated three-dimensional digital environment. The virtual scene may include objects, characters, and landscapes designed to create an immersive gaming experience for the player. The virtual scene may also include interactive elements that allow the player to interact with the environment and affect the game's outcome. The virtual scene may be rendered in real-time using advanced graphics processing techniques, allowing the player to move and interact with the environment seamlessly and responsively.

At operation 204, the process 200 includes generating a scanning query toward a first target area of the virtual scene. An example of a scanning query is a segment cast that can return all the geometric information in a specific region of the virtual scene.

In some cases, the system can generate multiple segment casts, each cast towards a respective target area within the virtual scene. For instance, the first segment cast can capture geometric features corresponding to the region parallel to the head of the player avatar while standing straight, and the second segment cast can capture features corresponding to the ground level when the avatar is on the ground. The example system can execute these two threads in parallel to enable faster processing.

In some cases, the system can generate multiple scanning queries, each returning geometric feature data associated with a respective region of the virtual scene. For instance, a first scanning query can return geometric feature data associated with a region parallel to the player avatar's head. A second scanning query can return geometric feature data associated with a region parallel to the player's avatar feet. These scanning queries can return geometric feature data representing the detected presence of a particular geometric shape and/or a transition in the ground level of the virtual scene.

In some cases, the scanning query includes a segment cast associated with a plurality of dimensions, and the first target area is determined based on the plurality of dimensions and with reference to a vantage point associated with the virtual scene. In some cases, the segment cast is a two-dimensional segment cast with a time dimension and a half axis dimension. In some cases, the segment cast is a three-dimensional segment cast with a time dimension, a half axis dimension, and a height extrusion axis dimension.

In some cases, the first target area associated with the scanning query includes at least a portion of a line of sight of the avatar when the line of sight is substantially parallel to the ground level. In some cases, the first target area associated with the scanning query includes at least a portion of the ground level of the virtual scene.

In some cases, the scanning query includes at least one of a ray cast, a collection of ray casts, or a segment cast. A ray cast may involve casting a straight line in a particular direction from an initial point in the virtual scene. A collection of ray casts may involve casting multiple rays in different directions from an initial point in the virtual scene. A segment cast may include scanning a particular region of the virtual scene for all geometric information.

In some cases, generating the scanning query may include at least one of a voxel traversal or a hierarchical traversal. Voxel traversal may involve dividing the virtual scene into a grid of voxels (3D pixels) and scanning each voxel to detect any objects or obstacles present in that location. Hierarchical traversal may involve dividing the virtual scene into smaller regions and scanning each region for objects or obstacles

At operation 206, the process 200 includes determining a first geometric feature based on the output data returned by the scanning query. An example of a geometric feature that the scanning query could return is the height of the terrain or objects within the scanned region. An example of a geometric feature that the scanning query could return is the orientation of a surface, such as whether the described surface is sloped or flat. The scanning query could also return information related to the texture or material properties of one or more objects in the scanned region. In some cases, the first geometric feature is determined based on the intersection of a plane associated with a segment cast and an object in the first target area.

At operation 208, the process 200 includes determining whether the first geometric feature represents a predefined scene feature. Examples of predefined scene features include obstacles (e.g., obstacles in a region within the predicted trajectory of the avatar that is substantially aligned with a head of the avatar while the avatar is in an upright position) and transitions in a ground level of the virtual scene.

Geometric features in a virtual scene that indicate a transition in the ground level typically may involve changes in elevation or curvature of the terrain. For example, if the ground level in front of the avatar suddenly becomes significantly steeper, it may indicate the presence of a transition in the ground level. Such steepness change detections can be used to detect hills or slopes that the avatar should optimally climb or descend. As another example, if the ground level in front of the avatar changes in curvature, it may indicate the presence of a transition in the ground level. Such curvature change detections can detect bumps or ridges in the terrain that the avatar should optimally navigate.

In some cases, if the virtual scene includes a staircase that the avatar must ascend or descend, it may indicate the presence of a transition in the ground level. Such staircase detections can be used to detect indoor or outdoor staircases that the avatar should optimally climb or descend.

In some cases, the presence of a transition in the ground level may be indicated by visual cues such as a change in texture or color of the terrain. Such visual cues can be used to detect changes in terrain such as rocky areas or sand dunes.

At operation 210, the process 200 includes controlling the player avatar without any modifications based on (e.g., in response to) determining that the first geometric feature does not represent a predefined scene feature. In some cases, based on (e.g., in response to) determining that the first geometric feature does not represent a predefined scene feature, the system controls the player avatar based on player input without any modifications. Accordingly, the system may skip modifying the avatar actions/movements based on mapping predefined scene features to recommended actions.

At operation 212, the process 200 includes controlling the player avatar by modifying the avatar movements based on a recommended action mapped to the predefined scene feature. In some cases, based on (e.g., in response to) determining that the first geometric feature represents a predefined scene feature, the system modifies the actions of the avatar based on the recommended action associated with the predefined scene feature.

For example, if the predefined scene feature is a head-level obstacle, the system may control the avatar by causing the avatar to be in a lowered head (e.g., crouch) position even if the player does not provide input data (e.g., does not perform actions) configured to lower the avatar's head. As another example, if the predefined scene feature is a ground-level transition (e.g., a hill, a downhill, a hole, a staircase, and/or the like), the system may control the avatar by causing the avatar to adjust its movement direction after the transition to reduce or eliminate the effect of the transition on the avatar's direction.

In some cases, the first target area comprises at least a portion of a line of sight of the avatar when the line of sight is substantially parallel to the ground level. In some cases, the first predefined scene feature includes a first obstacle. In some cases, controlling the avatar based on the recommended action includes automatically causing the avatar to transition to a posture configured to avoid collision between the avatar and the first obstacle. In some cases, the first target area is determined based on a region that includes at least a portion of the ground level of the virtual scene. In some cases, the first predefined scene feature includes a first ground-level transition. In some cases, controlling the avatar based on the recommended action includes automatically adjusting the orientation of the avatar (e.g., to adjust the effect of the transition on the direction of movement associated with the avatar).

FIGS. 3A-3B provide an operational example of detecting a scene feature 306 that is a head-level obstacle. As depicted in FIG. 3A, while the avatar 302 is moving in the virtual scene 300, the system generates a segment cast 304. At the time associated with FIG. 3A, the segment cast 304 does not capture geometric feature data that represents the presence of scene feature 306. However, at the time associated with FIG. 3B, the segment cast 304 captures the geometric feature 308, representing the presence of the scene feature 306. Accordingly, detection of scene feature 306 causes the system to control the avatar 302 by automatically lowering the avatar's head.

FIG. 4 provides an operational example of detecting a scene feature 406 that corresponds to a ramp. A ramp may be a type of transition in the ground level of the virtual scene 400. As depicted in FIG. 4, the segment cast 404 captures the geometric feature 408, representing the presence of the scene feature 406 in the virtual scene 400. Accordingly, the system controls the avatar 402 to adjust the direction of the avatar 402 after the avatar jumps through the ramp.

FIG. 5 provides an operational example of detecting a scene feature 506 that corresponds to a stairstep. A stairstep may be a type of transition in the ground level of the virtual scene 500. As depicted in FIG. 5, the segment cast 504 captures the geometric feature 508, representing the presence of the scene feature 506 in the virtual scene 500. Accordingly, the system controls the avatar 502 to adjust the direction of the avatar 502 after the avatar goes up the stairstep.

FIG. 6 provides an operational example of detecting a scene feature 606 that corresponds to a skating bowl. A skating bowl may be a type of transition in the ground level of the virtual scene 600. As depicted in FIG. 6, the segment cast 604 captures the geometric feature 608, representing the presence of the scene feature 606 in the virtual scene 600. Accordingly, the system controls the avatar 602 to adjust the direction of the avatar 602 after the avatar goes down the skating bowl.

FIG. 7 illustrates a block diagram of example game system(s) 110 that may provide assisted gameplay in accordance with examples of the disclosure. The game system(s) 110 may include one or more processor(s) 700, one or more input/output (I/O) interface(s) 702, one or more network interface(s) 704, one or more storage interface(s) 706, and computer-readable media 708.

In some implementations, the processor(s) 700 may include a central processing unit (CPU), a graphics processing unit (GPU), both CPU and GPU, a microprocessor, a digital signal processor or other processing units or components known in the art. Alternatively, or in addition, the functionally described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that may be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip system(s) (SOCs), complex programmable logic devices (CPLDs), etc. Additionally, each of the processor(s)