Video playing method, device, equipment and medium
A video playback and video technology, applied in image communication, selective content distribution, electrical components, etc., can solve problems such as video content occlusion, achieve intelligent dynamic layout, and avoid occlusion effects
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0027] FIG. 1a is a flowchart of a video playback method provided in Embodiment 1 of the present invention. This embodiment can be applied to
[0029] Wherein, the current video can be any playable video.
[0030] The video image is a frame of image currently playing in the current video.
[0031] Referring to Figure 1b, component 101 refers to an identification of an operation. The component 101 here is the standard for operating on the video.
[0032] The number of the components may be one, two or more. The component is located on the same display as the current video
[0033] The current location of the component is the current location of the component. The display size of the component is preset, specifically
[0034] The position adjustment information of the component refers to the position adjustment information of the component that avoids the occlusion of the video content by the component.
[0036] According to the currently played video image, the current position ...
Embodiment 2
[0070] FIG. 2 is a flowchart of a video playback method provided in Embodiment 2 of the present invention. This embodiment is based on the above
[0071] S210, the training image is marked with the main body region, and the main body marked image is generated.
[0072] S220, generate positive samples and negative samples according to the subject labeling image.
[0074] Specifically, the positive sample includes: a subject annotation image, the current position of the component, and the display of the component
[0075] S230, using positive samples and negative samples to train the initial adjustment model to obtain an occlusion adjustment model.
[0076] S240, the currently playing video image is projected on the canvas to generate the current video image.
[0077] Wherein, the current video image only includes the currently played video image, does not include component information.
[0082] The technical solution of the embodiment of the present invention is to learn a position...
Embodiment 3
[0087] Referring to Fig. 3b, a detailed description of how the middle layer calculates and moves the components is as follows:
[0088] Use the video tag on the web page to obtain the current video.
[0089] For example, the content of the video resource is obtained according to the address of the current video resource in the video.
[0093] According to the position adjustment information of the component and the position adjustment condition of the component, the position adjustment information of the component is determined.
[0094] Wherein, the position adjustment condition of the component includes whether the component is allowed to be adjusted, and restricts the adjustment direction.
[0096] Among them, the occlusion adjustment model is obtained by supervised regression training through machine learning. The training process is specifically:
[0099] The technical solution of the embodiment of the present invention, through the interaction between the video content and t...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


