Selective Video Encoding for Changing Viewpoints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding methods lack the ability to determine on the decoding side whether a desired image can be reproduced without using corresponding image data from a certain camera, leading to inefficient encoding of all video camera images.
Innovation Solution
A video encoding and decoding technique that assigns images to Groups of Pictures (GOPs) and determines whether each image needs to be encoded, allowing for selective encoding and decoding based on image generation methods, indicating whether encoded data can be reproduced without using specific camera images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If all video camera images are encoded using conventional methods, then complete image data is preserved, but encoding efficiency is reduced and data volume increases
Solution Approach 1:
The patent extracts only the essential image data that cannot be reproduced from other cameras' images. By identifying and encoding only the unique contribution of each camera, the system removes redundant data while preserving the ability to reconstruct any desired viewpoint, thereby improving encoding efficiency and reducing data volume.
Solution Approach 2:
The patent changes the encoding parameter from encoding all camera images to encoding only the irreducible image data. This parameter change is achieved by introducing a determination mechanism that identifies which image data can be reproduced from other cameras and should therefore be excluded from encoding, optimizing the encoding process.
2Reliability
If image data from all cameras is encoded, then image reproduction quality is maintained, but processing time and computational resources increase
Solution Approach 1:
The patent extracts and identifies image data that can be reproduced from other cameras' images before the encoding process. By determining in advance which data is redundant, the system avoids unnecessary processing of redundant data while ensuring that only essential data is encoded, thus maintaining reproduction quality while reducing processing time.
Solution Approach 2:
The patent performs preliminary determination of which image data can be reproduced from other cameras before the actual encoding process. This preliminary action allows the system to prepare the encoding process by identifying redundant data in advance, preventing wasted computational resources and processing time during the encoding phase.
3Productivity
If selective encoding is implemented without determination capability, then encoding efficiency improves, but image reproduction accuracy deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the determination result of whether an image can be reproduced from other cameras feeds into the encoding decision. This feedback loop ensures that encoding is selectively applied only when necessary for accurate reproduction, maintaining both encoding efficiency and reproduction accuracy through informed selective encoding.
Solution Approach 2:
The patent changes the encoding parameter based on the determination result. When the determination indicates that an image can be reproduced from other cameras, the encoding parameter is set to skip encoding; when it cannot be reproduced, encoding is performed. This dynamic parameter change ensures accurate reproduction while optimizing encoding efficiency.
Data Source
AI summary
A video encoding method for assigning a plurality of images to a plurality of GOPs and encoding images belonging to the GOPs as a video image. The method includes determining whether each image belonging to each GOP is to be encoded; encoding GOP encoding/non-encoding data for indicating whether encoded data of the image belonging to the relevant GOP is output; and encoding the image belonging to the relevant GOP when the encoded data of the image is output. Typically, it is determined whether an image generated by using one or more other GOPs without decoding the encoded data of the relevant GOP is closer to an original image of the relevant image in comparison with an image obtained by decoding the encoded data, so as to determine whether the image belonging to the relevant GOP is to be encoded.