- Number
- 10080066
- Published
- 2018-09-18
- Filed
- 2016-08-23
- Assignee
- Disney Enterprises, Inc.
- Inventors
- Rao; Cen et al.
- CPC
- G06T7/194; G06T19/006; H04N21/2187; H04N21/8146; G06T7/75; H04N21/23418
- Verdict
- Set aside video watermark/ad-mark insertion plumbing
- Source
- Google Patents · FreePatentsOnline
Abstract
One embodiment provides a method for rendering a virtual mark on a video frame, including: receiving the video frame, captured by a camera, comprising a scene; obtaining a three-dimensional model of a mark in the scene; obtaining a foreground mask of the video frame; denoting a predetermined region corresponding to a mark mask, wherein the mark mask is an image projection of the mark on the video frame; updating the mark mask to exclude foreground regions, wherein the updating comprises: partitioning the predetermined region into a plurality of segments; and filtering out, from the mark mask, foreground pixels based on the plurality of segments and the foreground mask; and rendering, on the video frame, the virtual mark using the updated mark mask. Other aspects are described and claimed.
Background
BACKGROUND(1) Many users watch sports content (e.g., games, sporting events, etc.) at a remote location (e.g., home, a bar, at a friend's house, etc.), rather than the user being physically present at the sporting event. To view the sports content, users use an information handling device (e.g., a television, smart television, personal computer, mobile device, etc.). As technological advancement increases, the amount of information available to the viewer while watching the sports content on an information handling device also increases. Specifically, augmented reality techniques allow for insertion of enhancements (i.e., virtual elements such as graphics) into the broadcast video in a perspective that is consistent with the broadcast camera's perspective.(2) For example, a user watching the game on an information handling device, which allows for user interaction, can select a player and learn different statistics associated with that player. Some devices may insert statistics or other game related information onto the sports content or may alternatively direct the user to another information location (e.g., website, database, etc.). As another example, commentators may provide graphical inputs explaining a play occurring during the event. While explaining the play, the broadcaster may provide additional content, for example, by drawing circles or lines on a screen, which is then visible to the viewer of the content. Some information may also be inserted automatically by a s
Claims
1. A method for rendering a virtual mark on a video frame, comprising: receiving the video frame, captured by a camera, comprising a scene; obtaining a three-dimensional model of a mark in the scene, wherein the three-dimensional model identifies locations of objects within the scene; obtaining a foreground mask of the video frame; denoting a predetermined region corresponding to a mark mask, wherein the mark mask is an image projection of the mark on the video frame and wherein the denoting comprises projecting the three-dimensional model onto the video frame to identify a location of the mark; updating the mark mask to exclude foreground regions from the mark mask, wherein the updating comprises: identifying, based on the foreground mask, pixels of the mark mask attributable to at least one foreground object and excluding the identified pixels; and reallocating a pixel incorrectly attributed to at least one of a background object and a foreground object by partitioning the predetermined region into a plurality of segments and analyzing the plurality of segments using the foreground mask; and filtering out, from the mark mask, foreground pixels; and rendering, on the video frame, the virtual mark using the updated mark mask.
11. A system for rendering a virtual mark on a video frame, comprising: an electronic device that includes a processor; a memory device that stores instructions executable by the processor to: receive the video frame, captured by a camera, comprising a scene; obtain a three-dimensional model of a mark in the scene, wherein the three-dimensional model identifies locations of objects within the scene; obtain a foreground mask of the video frame; denote a predetermined region corresponding to a mark mask, wherein the mark mask is an image projection of the mark on the video frame and wherein the denoting comprises projecting the three-dimensional model onto the video frame to identify a location of the mark; update the mark mask to exclude foreground regions from the mark mask, wherein to update comprises: identifying, based on the foreground mask, pixels of the mark mask attributable to at least one foreground object and excluding the identified pixels; and reallocating a pixel incorrectly attributed to at least one of a background object and a foreground object by partitioning the predetermined region into a plurality of segments and analyzing the plurality of segments using the foreground mask; and filtering out, from the mark mask, foreground pixels; and render, on the video frame, the virtual mark using the updated mark mask.
20. A program product, comprising: a computer readable storage device having code embodied therewith, the code being executable by a processor and comprising: code that receives the video frame, captured by a camera, comprising a scene; code that obtains a three-dimensional model of a mark in the scene, wherein the three-dimensional model identifies locations of objects within the scene; code that obtains a foreground mask of the video frame; code that denotes a predetermined region corresponding to a mark mask, wherein the mark mask is an image projection of the mark on the video frame and wherein the denoting comprises projecting the three-dimensional model onto the video frame to identify a location of the mark; code that updates the mark mask to exclude foreground regions from the mark mask, wherein the code that updates comprises: code that identifies, based on the foreground mask, pixels of the mark mask attributable to at least one foreground object and excluding the identified pixels; and code that reallocates a pixel incorrectly attributed to at least one of a background object and a foreground object by partitioning the predetermined region into a plurality of segments and analyzing the plurality of segments using the foreground mask; and code that filters out, from the mark mask, foreground pixels; and code that renders, on the video frame, the virtual mark using the updated mark mask.