Image Stabilization Subject Blur Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image stabilization technologies in image capture devices fail to effectively differentiate between hand shake and subject shake, leading to incomplete blur correction that does not align with the user's intentions, and cause unnatural shifts in video when switching between correction modes.
Innovation Solution
An image stabilization apparatus and method that detects specific subjects and background motion, estimates the target of interest using subject and camera information, and combines blur correction amounts for hand shake and subject shake to generate a final correction amount based on user intent, allowing for dynamic adjustment of correction ratios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If face detection-based binary switching is used to determine shake correction target, then the correction mode can be simplified, but it causes unnatural video shifts and cannot adapt to user intentions
Solution Approach 1:
The patent applies dynamics by transitioning from static binary switching to dynamic continuous adjustment. The system dynamically determines the degree of interest in the subject based on detected subject motion, and continuously adjusts the correction ratio between subject shake and hand shake correction accordingly. This allows the correction behavior to adapt flexibly to changing scene conditions and user intentions rather than being locked into fixed modes.
Solution Approach 2:
The patent changes the parameter of correction ratio from discrete binary states to continuous values. By introducing a continuous parameter (degree of interest) that ranges from 0 to 1, the system can smoothly adjust the weighting between subject shake correction and hand shake correction. This parameter change enables nuanced control over correction behavior to match varying user intentions across different shooting scenarios.
2Manufacturing precision
If subject shake correction is applied when subject is detected, then subject blur can be corrected, but it may correct wrong motion when background is actually the target of interest
Solution Approach 1:
The patent implements feedback by using detected subject motion information to continuously adjust the correction strategy. The system detects subject motion, determines the degree of interest based on this motion, and feeds this information back to adjust the correction ratio. This closed-loop feedback mechanism ensures that correction behavior aligns with actual scene conditions and user intentions, preventing incorrect correction when the background is the intended target.
Solution Approach 2:
The system dynamically adjusts the correction approach based on real-time subject motion detection. Rather than applying fixed correction rules, the system continuously adapts the correction ratio between subject shake and hand shake correction based on the determined degree of interest. This dynamic adaptation allows the system to switch smoothly between correcting subject blur and hand shake based on what the user actually intends to capture.
3Device complexity
If binary mode switching is used for shake correction, then system complexity is reduced, but it causes unnatural shifts in video during mode transitions
Solution Approach 1:
The patent ensures continuity by maintaining smooth transitions between correction modes through continuous parameter adjustment. Instead of abruptly switching between discrete correction modes, the system continuously adjusts the correction ratio based on the degree of interest in the subject. This continuous adjustment eliminates sudden jumps in correction behavior, ensuring smooth and natural video playback without unnatural shifts during mode transitions.
4Stability of the object's composition
If only hand shake correction is applied, then entire screen stability is improved, but subject blur cannot be corrected when subject moves
Solution Approach 1:
The patent applies segmentation by separating the correction of subject shake from hand shake correction. The system detects and separates subject motion from camera shake, then applies appropriate correction to each component independently. This segmentation allows the system to correct subject blur when the subject moves while maintaining overall image stability, by selectively applying correction to the subject portion rather than treating all motion uniformly.
Data Source
AI summary
An apparatus includes a subject detection unit configured to detect a specific subject in an input image, an acquisition unit configured to acquire camera information, an estimation unit configured to estimate a target of interest in an image using the subject information and the camera information, a first motion detection unit configured to detect background motion and subject motion, a conversion unit configured to convert the detected background motion and the detected subject motion to a first blur correction amount for correcting background blur and a second blur correction amount for correcting subject blur, respectively, and a correction amount calculation unit configured to, based on the target of interest, combine the first blur correction amount and the second blur correction amount and generate a final blur correction amount.