3DoF+ 360 Video Location-Based Asymmetric Down-Sampling for View Synthesis to Immersive VR Video Streaming.

Area of Science:

• Computer Science
• Multimedia Systems
• Virtual Reality

Background:

• Virtual reality (VR) adoption is growing, increasing demand for immersive 360 video content.
• Processing and streaming 360 videos, especially stereoscopic content requiring depth information, demands significant computational resources and bandwidth.
• Current methods face challenges in delivering high-quality immersive experiences efficiently.

Purpose of the Study:

• To propose an efficient streaming method for high-quality 360 videos.
• To reduce bandwidth consumption and computational load for 3DoF+ (3 Degrees of Freedom plus) 360 video streaming.
• To enable immersive stereoscopic 360 content delivery.

Main Methods:

• Analyzed the relationship between bitrate and peak signal-to-noise ratio (PSNR) to determine optimal down-sampling ratios and quantization parameters.
• Utilized High-Efficiency Video Coding (HEVC) for encoding and decoding 360 videos.
• Employed a view synthesizer to generate intermediate views for an enhanced user experience.

Main Results:

• Identified optimal down-sampling ratios and quantization parameters for efficient 360 video streaming.
• Demonstrated significant bandwidth reduction for 3DoF+ 360 videos.
• Achieved high-quality video synthesis for immersive stereoscopic viewing.

Conclusions:

• The proposed method effectively reduces bandwidth requirements for streaming high-quality 360 videos.
• Optimized HEVC encoding parameters and view synthesis contribute to efficient and immersive VR experiences.
• This approach facilitates the wider adoption of stereoscopic 360 content in virtual reality applications.