Joint-Conditioned Image Generation With Occlusion Consistency Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image generation systems using generative AI often produce images with inconsistencies between input joint information and generated images, such as missing or added body parts, which can hinder effective learning and application in tasks like machine learning.
Innovation Solution
An image processing apparatus and method that includes joint information input, generation, detection, occlusion determination, and consistency determination units to ensure alignment between input joint information and generated images, using techniques like deep learning and three-dimensional surface models to correct occlusions and inconsistencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If generative AI is used to automatically generate images from joint information, then image generation efficiency is improved, but consistency between input joint information and generated images deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the generated image is fed back into the system for joint detection and occlusion determination. The detection unit detects joints from the generated image, and the occlusion determination unit determines occlusion states based on detected joints and input joint information. This feedback loop enables consistency determination and allows for iterative improvement of image generation quality.
Solution Approach 2:
The patent performs preliminary actions by detecting joints from the generated image before final consistency determination. The detection unit detects joints in advance, and the occlusion determination unit determines occlusion states preliminarily, allowing the consistency determination unit to make informed decisions about image quality and consistency with input joint information.
2Reliability
If joint detection and occlusion determination are performed to improve consistency, then image quality is improved, but processing time increases
Solution Approach 1:
The patent segments the image processing task into distinct functional units: joint detection, occlusion determination, and consistency determination. Each unit performs a specific function independently, which allows for optimized processing and parallel execution. The detection unit handles joint detection, the occlusion determination unit handles occlusion analysis, and the consistency determination unit performs final consistency checking, improving overall efficiency.
Solution Approach 2:
The patent applies partial action by focusing the consistency determination on specific aspects such as joint consistency and occlusion state verification rather than performing a complete reanalysis of the entire image. This selective approach maintains image quality while reducing processing time by concentrating computational resources on the most critical consistency checks.
Data Source
AI summary
An image processing apparatus includes an input unit that inputs joint information of a generation target, a generation unit that generates an image of the generation target based on the joint information, a detection unit that detects a joint from a generated image generated by the generation unit, an occlusion determination unit that determines an occlusion state of the joint in the generated image, and a consistency determination unit that determines consistency between the joint information and the generated image based on a detection result of the joint and the occlusion state of the joint.