Video Special Effect Generation via Subject Region Overlay

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video special effect technologies require professional post-processing and complex tools, making it difficult to implement and preview special effects during video recording, with a high learning cost and professional requirement for personnel.

Innovation Solution

A method and device that automatically identifies and isolates a target object in video frames using image segmentation, overlaying the subject region on other frames to generate a doppelgänger effect without needing complex post-production tools, allowing for direct special effect generation during photography.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If professional post-processing tools and techniques are used to create video special effects, then the quality and variety of special effects are improved, but the device complexity and learning cost increase significantly

Engineering Contradiction:
Improvespecial effect qualityVSAvoidpost-processing tool complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The terminal device performs automatic subject region identification and segmentation using its own processing capabilities, eliminating the need for external professional post-processing tools. The device identifies the target object, segments the subject region, and generates the doppelgänger effect independently, making the system self-sufficient and reducing external tool requirements.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent extracts the key function of subject region identification and segmentation from complex post-processing software and integrates it directly into the terminal device's processing pipeline. By taking out the essential functionality and embedding it in the recording device, the system eliminates the need for separate professional editing tools while maintaining special effect quality.

Inventive Principle:
Principle #2Taking out (Extraction)

2Manufacturing precision

If professional post-processing is required to generate video special effects, then the special effect production quality is improved, but the ease of operation and accessibility are reduced

Engineering Contradiction:
Improvespecial effect production qualityVSAvoidspecial effect generation ease
Core Design Contradiction:
Manufacturing precisionVSEase of operation

Solution Approach 1:

The patent merges the special effect generation function with the video recording function by integrating subject region identification, segmentation, and overlay processing directly into the terminal device. This combination allows users to capture and generate doppelgänger effects in one step during recording, eliminating the need for separate post-processing operations and significantly improving ease of operation.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system performs subject region identification and segmentation in advance during the video recording process itself, rather than requiring post-processing. By completing the segmentation action preliminarily while recording, the device prepares the processed video frames for immediate overlay, enabling real-time special effect generation without requiring users to perform complex editing operations later.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If real-time special effect preview is implemented during video recording, then the user experience is improved, but the processing speed and system resource consumption increase

Engineering Contradiction:
Improvereal-time preview capabilityVSAvoidprocessing speed
Core Design Contradiction:
Ease of operationVSSpeed

Solution Approach 1:

The device performs subject region identification and segmentation preliminarily during video recording, generating processed video frames in advance. This preliminary processing enables real-time preview of the doppelgänger effect during recording without requiring heavy post-processing computation, as the segmentation work is already completed when the frames are generated for overlay.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the video processing into distinct stages: subject region identification, segmentation to extract the target object, and overlay generation. By segmenting the processing pipeline, the system can perform identification and segmentation in real-time during recording for preview purposes, while the final composite generation can be optimized separately, balancing processing speed and preview capability.

Inventive Principle:
Principle #1Segmentation

4Manufacturing precision

If complex post-production techniques are used to create doppelgänger effects, then the visual quality is improved, but the productivity and efficiency of video production are reduced

Engineering Contradiction:
Improvevisual qualityVSAvoidvideo production efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent combines multiple post-production operations (subject identification, segmentation, frame overlay, and alignment) into a single integrated process that occurs during video recording. By merging these functions into the terminal device's processing pipeline, the system maintains high visual quality through precise subject region extraction and overlay while dramatically improving productivity by eliminating separate post-production steps.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The device performs all necessary processing steps preliminarily during recording, including subject region identification, segmentation, and preparation of overlay frames. By completing these actions in advance during the recording phase, the system eliminates time-consuming post-production operations while maintaining visual quality, thereby significantly improving overall video production efficiency.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3945490B1Method and device for processing video, and storage medium
Publication Date: 2024.08.21 BEIJING XIAOMI PINECONE ELECTRONICS CO LTD
  • EP3945490B1 patent drawingFigure 1~2
  • EP3945490B1 patent drawingFigure 3~4
  • EP3945490B1 patent drawingFigure 5

AI summary

A method is applied to a terminal (800), and includes: determining (S101) a subject region of a video frame in a video, in which a target object is located, and a background region, the background region being a region of the video frame other than the subject region; and overlaying (S102) the subject region in at least one of a first video frame having the target object on at least one of a second video frame having the target object, generating a special effect frame including at least two subject regions in each of which the target object is located.