Face Detection Algorithm for Automatic Group Photo Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Photographers face challenges in capturing images of living subjects, particularly groups, as they tend to move and pose incorrectly, leading to the need for multiple shots and wasteful photo processing, with existing solutions requiring manual editing and storage of imperfect images.

Innovation Solution

A photographic apparatus and method that uses computer-implemented algorithms to detect and ensure subjects are in a predetermined position before capturing an image, combining face detection and analysis to automatically record images when all subjects are properly positioned, replacing traditional countdown timers and manual editing processes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If multiple shots are taken to ensure all subjects are properly positioned, then the quality of the final photograph is improved, but the time required and number of images processed increases

Engineering Contradiction:

Improvephotograph qualityVSAvoidtime to capture and process images

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of captured images to detect subject positions, face orientations, and eye states before final selection. This preliminary action allows the camera to automatically determine which images contain properly positioned subjects without requiring manual review of each shot, thereby reducing both time and processing effort while maintaining high photograph quality

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system provides feedback to the user by displaying thumbnails of captured images with indicators showing which subjects are properly positioned (facing camera, eyes open). This feedback mechanism allows the user to quickly assess which images meet the criteria for a perfect group photograph, enabling rapid selection without thorough manual inspection of each image

Inventive Principle:

Principle #23Feedback

2Productivity

If burst mode is used to capture multiple images rapidly, then the speed of capturing is improved, but the internal memory fills quickly and transfer to external memory becomes a bottleneck

Engineering Contradiction:

Improveimage capture speedVSAvoidmemory capacity

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The system performs preliminary analysis of captured images to identify those containing properly positioned subjects before transferring them to external memory. By pre-screening images and selecting only those that meet the criteria, the system reduces the total number of images that need to be stored and transferred, thereby extending the effective burst mode duration and reducing the burden on internal and external memory capacity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system extracts and selects only the relevant images from the burst sequence - specifically those containing all subjects in proper positions - and transfers only these selected images to external memory. This extraction process eliminates the need to store and transfer unnecessary imperfect images, optimizing memory usage and transfer efficiency

Inventive Principle:

Principle #2Taking out (Extraction)

3Manufacturing precision

If manual editing is performed to correct subject positions in photographs, then the quality of the final image is improved, but the complexity of the process increases

Engineering Contradiction:

Improvephotograph qualityVSAvoidediting process complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system performs automatic analysis and selection of images with properly positioned subjects, eliminating the need for manual editing to correct subject positions. The camera's image processing unit automatically detects face orientations, eye states, and subject positions, and selects the best images without requiring user intervention or post-processing editing, thereby maintaining high photograph quality while simplifying the overall process

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces manual editing operations with automated electronic image analysis. Instead of requiring users to manually review and potentially edit images to correct subject positioning, the system uses electronic algorithms to automatically detect and select images where all subjects are properly positioned, substituting mechanical manual editing with automated electronic processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8059163B2Digital photography using electronic image analysis to identify faces of faces of known subjects in predetermined position

Publication Date: 2011.11.15 SONY INTERACTIVE ENTERTAINMENT LLC

AI summary

A method, apparatus and computer readable medium for use in photography are disclosed. A first electronic image analysis is performed on a potential image to determine whether any suitable subjects are within a frame of the potential image. If any suitable subjects are found in the frame, a second electronic image analysis is performed to determine whether the suitable subjects are in a predetermined position. The potential image is recorded if the second image analysis determines that the suitable subjects are in the predetermined position.