Depth Image Void Correction for Accurate Hand Joint Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current gesture recognition systems suffer from inaccurate interaction positions due to abnormal detection of joint points in hand gestures, leading to poor user experience.

Innovation Solution

A method for correcting void positions in depth images by aligning visible light and depth image sequences, using a hand region acquisition model and joint point detection model to determine and correct voids based on previous frames, and applying weighted averages and morphological operations to improve joint point accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If gesture recognition is performed using depth images, then interaction capability is enabled, but joint point detection accuracy deteriorates due to voids in depth data

Engineering Contradiction:
Improvegesture interaction capabilityVSAvoidjoint point detection accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary detection of void positions in depth images before joint point detection. By identifying void regions in advance and filling them using information from previous frames and spatial interpolation, the system ensures that joint point detection operates on corrected depth data, thereby improving detection accuracy while maintaining gesture interaction capability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary processing step that uses visible light images and historical depth frames as mediators to fill voids in current depth images. This intermediary process combines temporal information (previous frames) and spatial information (visible light images) to reconstruct missing depth data, enabling accurate joint point detection despite depth image voids

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If void correction using previous frames is applied, then depth image quality is improved, but processing time increases

Engineering Contradiction:
Improvedepth image qualityVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies void correction selectively only to regions identified as voids in the depth image, rather than processing the entire image. By detecting void positions and applying correction algorithms only to these specific regions using local spatial interpolation and temporal fusion, the system improves depth quality where needed while minimizing unnecessary processing overhead

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent performs partial correction by focusing computational resources on void regions rather than processing the complete depth image. The correction uses a limited number of previous frames (N frames) and applies morphological operations only where needed, achieving sufficient depth quality improvement without excessive processing time

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4697266A1Void correction method, electronic device, and computer-readable medium
Publication Date: 2026.02.18 ZTE CORP
  • EP4697266A1 patent drawingFigure 1
  • EP4697266A1 patent drawingFigure 2~3
  • EP4697266A1 patent drawingFigure 4~5

AI summary

Provided in the present application are a void correction method, an electronic device, and a computer-readable medium. The void correction method comprises: acquiring a visible image sequence and a first depth image sequence having the same acquisition frame rate; for each first depth image in the first depth image sequence, on the basis of the first depth image and a visible image in the visible image sequence meeting a preset condition, determining whether the first depth image is a depth image having a void, wherein the visible image meeting the preset condition is a visible image of which the position corresponds to that of the first depth image; when it is determined that the first depth image is a depth image having a void, determining void position information of the first depth image; and, on the basis of N first depth images ahead of the first depth image, correcting the depth information of the void corresponding to the void position information, N being 1 or 2.