3D Rendered Image Encoding With Depth-Layer Object Tagging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for encoding 3D rendered images are inefficient and time-consuming, particularly in the process of motion estimation and compression, which can be enhanced by attribute tagging objects generated by a 3D engine.

Innovation Solution

Implementing a system that segments 3D data into layers and computes deformable objects using deformer parameters, eliminating the need for motion estimation by tagging object depth relative to a background depth.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional motion estimation and compression methods are used for encoding 3D rendered images, then encoding can be performed, but the process is time-consuming and bandwidth-intensive

Engineering Contradiction:
Improveencoding speedVSAvoidtransmission time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent segments the 3D scene into multiple depth layers (foreground objects, midground objects, background) and processes each layer separately with different encoding parameters. This segmentation allows the encoder to focus computational resources on moving objects in the foreground while using more efficient compression for static background elements, thereby reducing overall encoding time and bandwidth requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different encoding quality levels to different regions of the 3D scene based on their motion characteristics and depth. Moving objects in the foreground receive higher quality encoding with more precision, while static background objects use lower quality encoding. This local quality differentiation optimizes the balance between encoding speed and transmission quality, reducing unnecessary bandwidth consumption.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If motion estimation is performed to encode 3D rendered images, then encoding accuracy is improved, but processing time and computational resources increase

Engineering Contradiction:
Improvemotion estimation accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by pre-computing depth information and object motion vectors during the 3D rendering process itself, before the actual encoding begins. The 3D engine tags objects with depth attributes and motion information in advance, so that when encoding starts, this pre-computed data is immediately available. This eliminates the need for time-consuming motion estimation during encoding, as the motion data is already prepared from the rendering stage.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent merges the motion estimation function with the 3D rendering process. Instead of performing motion estimation as a separate post-processing step, the motion vector computation is integrated into the rendering pipeline where 3D objects are generated and tagged. This merging of functions allows motion data to be computed concurrently with rendering operations, eliminating sequential processing time and reducing overall computational burden.

Inventive Principle:
Principle #5Merging (Combining)

3Device complexity

If all pixels are encoded with the same quality, then encoding simplicity is maintained, but bandwidth efficiency decreases

Engineering Contradiction:
Improveencoding process simplicityVSAvoidbandwidth consumption
Core Design Contradiction:
Device complexityVSLoss of energy

Solution Approach 1:

The patent implements local quality encoding where different quality levels are applied to different spatial regions and depth layers. Foreground objects with significant motion receive high-quality encoding to maintain visual fidelity, while background objects with little or no motion use lower-quality encoding. This spatial and depth-based quality differentiation significantly reduces overall bandwidth consumption by avoiding unnecessary high-quality encoding of static background elements.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent segments the image data into multiple depth-based layers and processes each layer with appropriate quality settings. By separating the encoding process into distinct depth layers (foreground, midground, background), the system can apply optimized compression ratios to each layer, reducing total bandwidth requirements while maintaining encoding manageability through systematic layer processing.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3435672B1Encoding 3D rendered images by tagging objects
Publication Date: 2025.10.08 INTEL CORP
  • EP3435672B1 patent drawingFigure 1
  • EP3435672B1 patent drawingFigure 2A
  • EP3435672B1 patent drawingFigure 2B

AI summary

Image information is often transmitted from one electronic device to another. Such information is typically encoded and/or compressed to reduce the bandwidth required for transmission and/or to decrease the time necessary for transmission. Embodiments are directed to tagging objects or primitives with attribute tags to facilitate the encoding process. Other embodiments are directed to codecs running on hardware and/or software.