Non-Uniform Re-Sampling for 360 VR Projection Face Discontinuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The processing of omnidirectional image/video content for 360-degree virtual reality (VR) faces challenges in maintaining image quality due to high bitrate requirements and image content discontinuity at layout boundaries and face edges, leading to poor quality after compression.
Innovation Solution
A method involving re-sampling and padding techniques is employed to improve image quality, where projection faces are re-sampled with non-uniform mapping and padding regions are added to the 360 VR projection layout to reduce image content discontinuity and enhance compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If omnidirectional content is mapped onto projection faces using cubemap projection, then the 360-degree image/video content can be represented in a compact format, but image content discontinuity occurs at layout boundaries and face edges leading to poor image quality after compression
Solution Approach 1:
The patent applies preliminary action by performing re-sampling with non-uniform mapping before compression encoding. The re-sampling process pre-processes the projection faces to smooth out discontinuities at boundaries and edges, so that when compression is applied afterward, the image quality is preserved despite the compact packing format.
Solution Approach 2:
The patent implements local quality by applying non-uniform sampling density across different regions of the projection faces. Specifically, higher sampling density is applied to regions near boundaries and edges where discontinuities occur, while lower sampling density is used in interior regions. This localized approach maintains image quality at critical areas while improving overall compression efficiency.
2Manufacturing precision
If high resolution omnidirectional video (4K or higher) is delivered, then immersive visual experience is improved, but bitrate requirement increases significantly creating delivery bottleneck
Solution Approach 1:
The patent applies parameter changes by transforming the sampling density parameter across different spatial regions of the projection faces. By varying the sampling density from uniform to non-uniform, the patent achieves better compression characteristics that reduce bitrate requirements while maintaining perceptual image quality at 4K and higher resolutions.
Solution Approach 2:
The patent uses local quality by applying different sampling densities to different regions of the omnidirectional content. Critical regions near layout boundaries receive higher sampling density to preserve quality, while less critical interior regions use lower sampling density to reduce overall bitrate, achieving an optimal balance between quality and compression efficiency.
3Quantity of substance
If projection faces are packed in 360-degree VR projection layout, then compact representation is achieved, but artifacts are introduced by projection layout conversion causing image quality degradation
Solution Approach 1:
The patent applies preliminary action by performing re-sampling with non-uniform mapping before the projection layout conversion and compression processes. This pre-processing step anticipates and mitigates the artifacts that will be introduced during subsequent processing, thereby preserving image quality while maintaining compact representation.
Solution Approach 2:
The patent implements preliminary anti-action by applying non-uniform sampling that counteracts the expected artifacts from layout boundaries and edges. By pre-emphasizing sampling in regions where discontinuities will occur, the patent creates a compensatory effect that reduces visible artifacts after compression and layout conversion.
Data Source
AI summary
A video processing method includes: obtaining a plurality of projection faces from an omnidirectional content of a sphere, wherein the omnidirectional content of the sphere is mapped onto the projection faces via cubemap projection, and the projection faces comprise a first projection face; obtaining, by a re-sampling circuit, a first re-sampled projection face by re-sampling at least a portion of the first projection face through non-uniform mapping; generating a projection-based frame according to a projection layout of the cubemap projection, wherein the projection-based frame comprises the first re-sampled projection face packed in the projection layout; and encoding the projection-based frame to generate a part of a bitstream.