Spatial Layer Bit Allocation for Consistent Multi-Resolution Video Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding technologies face challenges in efficiently allocating bit rates across multiple spatial layers to ensure optimal video quality for different devices with varying capabilities, leading to potential compromises in user experience.
Innovation Solution
A method and system for determining a spatial rate factor based on transform coefficients and frame samples to allocate bit rates across spatial layers, using a spatial rate factor threshold and exponential moving averages to optimize bit rate allocation, ensuring high video quality across all spatial layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If bit rate is allocated uniformly across all spatial layers, then the encoding process is simple, but video quality deteriorates for layers requiring higher resolution
Solution Approach 1:
The patent applies local quality by allocating different bit rates to different spatial layers based on their specific requirements. Each spatial layer is assigned a customized bit rate calculated using the formula: bit rate for layer i = (spatial rate factor × average bit rate) / (1 + Σ(spatial rate factors)), where layers requiring higher resolution receive more bits while maintaining computational efficiency.
2Manufacturing precision
If bit rate is increased for higher spatial layers to maintain quality, then video quality improves for those layers, but total bit rate consumption increases
Solution Approach 1:
The patent uses parameter changes by dynamically calculating spatial rate factors based on transform coefficient statistics and frame samples. The bit rate allocation adapts to actual video content characteristics, allowing efficient resource utilization. The exponential moving average of spatial rate factors provides a smoothed parameter that balances quality requirements with bit rate constraints over time.
3Manufacturing precision
If bit rate allocation is optimized frame-by-frame, then video quality improves, but processing complexity and computational cost increase
Solution Approach 1:
The patent applies preliminary action by pre-calculating spatial rate factors based on transform coefficient statistics before final bit rate allocation. The exponential moving average is computed in advance to smooth out fluctuations, and these pre-computed factors are then used for efficient frame-by-frame allocation without requiring complex real-time optimization algorithms.
Solution Approach 2:
The patent implements feedback by using the exponential moving average of spatial rate factors, which continuously adapts to video content characteristics. The feedback mechanism allows the system to learn from previous frames and adjust bit rate allocation dynamically, improving quality while controlling complexity through the smoothing effect of the moving average.
Data Source
AI summary
A method includes receiving transform coefficients corresponding to a scaled video input signal, the scaled video input signal including a plurality of spatial layers that include a base layer. The method also includes determining a spatial rate factor based on a sample of frames from the scaled video input signal. The spatial rate factor defines a factor for bit rate allocation at each spatial layer of an encoded bit stream formed from the scaled video input signal. The spatial rate factor is represented by a difference between a rate of bits per transform coefficient of the base layer and an average rate of bits per transform coefficient. The method also includes reducing a distortion for the plurality of spatial layers by allocating a bit rate to each spatial layer based on the spatial rate factor and the sample of frames.