The proliferation of high-resolution displays and the shift towards cloud-native media workflows are driving an exponential increase in video data. Companies are seeking innovative solutions to manage escalating infrastructure costs, improve content delivery speeds, and meet sustainability goals. This technology offers a strategic advantage by enabling significant operational cost reductions and enhanced service quality in a highly competitive global streaming and data storage market.
Reduces Data Volume by up to 20%: By evaluating pixel-level similarity between reference images and selectively applying orthogonal transformation and quantization only to less critical areas of prediction residuals, this technology could efficiently reduce data volume, significantly lowering storage and bandwidth costs.
Maintains High Image Quality: Selective processing based on similarity evaluation minimizes quality degradation in visually important areas while maximizing overall encoding efficiency. This could deliver high-quality video without compromising user experience.
High Compatibility with Existing Systems: Defined as an image decoding apparatus and method, this technology could be integrated relatively easily as a software module into existing video distribution and storage infrastructures. It represents a stable right, having been granted patentability after standard prior art examination.
This patent protects an image decoding apparatus and method that improves encoding efficiency by selectively applying orthogonal transformation and quantization based on pixel-level similarity between reference images. The claims demonstrated clear novelty and inventiveness during examination, overcoming an initial rejection to secure a robust and stable right with low risk of future invalidation.
This patent primarily covers selective transformation and quantization in video decoding. Licensees could explore building additional IP around adaptive bitrate streaming protocols, AI-driven content analysis for dynamic compression, or hardware-accelerated implementations for specific edge devices without direct conflict.
Assuming an enterprise manages 100 PB of video data annually, with storage and distribution costs of ~$65K/PB/year (AI est.). If this technology reduces data volume by an average of 15%, an annual cost saving of 15 PB is expected. Calculation: 15 PB × ~$65K/PB = ~$1M/year (AI est.) in operational cost savings. This also contributes to reduced data center power consumption.
X: Data Reduction Efficiency
Y: Image Quality Preservation