Cloud VR Device MTP Latency Reduction via Orientation Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtual reality (VR) devices face issues such as high cost, low resolution, and motion-to-photon (MTP) latency, leading to discomfort like dizziness due to discrepancies between user head movement and visible VR video, particularly in wireless environments like WiFi or mobile networks.
Innovation Solution
A cloud VR device that includes a movement start detection unit, a Judder improvement unit, a video encoding processing unit, and a video image playback unit, which predict client orientation based on latency and user movement signals from tracking sensors, dynamically adjust latency, and perform foveated rendering to minimize MTP latency and improve image synchronization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If cloud VR system transmits user movement information through wireless network (WiFi/mobile network), then VR content accessibility is improved, but MTP latency increases causing physical discomfort
Solution Approach 1:
The system performs preliminary actions by detecting movement start through N DoF tracking sensors and predicting client orientation before the actual rendering process. The Judder improvement unit calculates predicted orientation based on angular velocity and MTP latency time, preparing the rendering target in advance to compensate for network transmission delays.
Solution Approach 2:
The system dynamically adjusts rendering parameters based on real-time movement detection. The video encoding processing unit encodes video images according to predicted orientation that changes dynamically with user movement, and the video image playback unit plays back the encoded images with adjusted timing to maintain synchronization with actual head orientation.
2Loss of time
If cloud VR system reduces MTP latency to 10-20 ms, then physical discomfort is minimized, but system complexity increases due to prediction and dynamic adjustment mechanisms
Solution Approach 1:
The cloud VR device is segmented into distinct functional units: movement start detection unit, Judder improvement unit, video encoding processing unit, and video image playback unit. Each unit handles a specific aspect of MTP latency reduction, making the complex system more manageable and maintainable while achieving the 10-20 ms latency target.
Solution Approach 2:
The Judder improvement unit acts as an intermediary between movement detection and video rendering. It receives angular velocity from tracking sensors, calculates predicted orientation considering MTP latency, and provides this predicted orientation to the video encoding processing unit, thereby mediating the timing discrepancy between user movement and visual feedback.
3Manufacturing precision
If system predicts client orientation based on latency, then image synchronization is improved, but calculation time and processing load increase
Solution Approach 1:
The system changes key parameters including MTP latency time, angular velocity, and predicted orientation to achieve synchronization. The Judder improvement unit dynamically adjusts these parameters based on real-time sensor data and network conditions, optimizing the balance between prediction accuracy and processing speed to maintain image synchronization within acceptable time frames.
Data Source
AI summary
Disclosed is a cloud VR device for MTP latency reduction. The cloud VR device includes a movement start detection unit detecting a movement start of a client virtual reality (VR) terminal, a Judder improvement unit predicting an orientation of the client VR terminal according to the movement start and providing the predicted orientation to a cloud VR server, a video encoding processing unit encoding a video image according to the predicted orientation through the cloud VR server and receiving the encoded video image, and a video image playback unit playing the encoded video image through the client virtual reality (VR) terminal.