Bodyprint Identity Recognition When Faces Are Obscured
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial recognition systems struggle to accurately identify individuals when their faces are not visible or of poor quality in video feeds, limiting the scenarios in which identity recognition can be performed.
Innovation Solution
The system utilizes physical characteristics such as the torso and clothing of a person, generating a 'bodyprint' to associate with their identity, allowing recognition even when facial features are obscured or not visible, using a multidimensional vector representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If facial recognition is used to identify individuals, then recognition speed is fast, but recognition accuracy deteriorates when the face is not visible or of poor quality
Solution Approach 1:
The system segments the identification task into two independent parts: facial recognition for known faces and bodyprint recognition for torso/clothing features. This allows the system to use facial recognition when available (high speed) and bodyprint recognition when facial recognition fails (maintaining accuracy), thereby resolving the contradiction between recognition accuracy and scenario range.
Solution Approach 2:
The system introduces bodyprint recognition as an intermediary mechanism that activates when facial recognition cannot provide accurate results. The bodyprint analysis of torso and clothing features serves as a mediator to maintain identification accuracy in scenarios where facial features are obscured, not visible, or of poor quality.
2Reliability
If only facial features are analyzed, then system complexity is low, but recognition reliability deteriorates in challenging lighting or obscured conditions
Solution Approach 1:
The system segments the analysis into separate modules: facial feature analysis and bodyprint analysis. Each module operates independently and can be activated based on conditions. This segmentation allows the system to maintain low complexity when only facial analysis is needed while adding bodyprint analysis only when reliability is compromised by lighting or obscuration conditions.
Solution Approach 2:
The system applies partial action by performing only the necessary analysis (facial recognition) when conditions permit, and excessive action by adding bodyprint analysis when facial recognition is insufficient. This ensures high reliability in challenging conditions without unnecessarily increasing system complexity in optimal conditions.
3Measurement precision
If bodyprint analysis is added to facial recognition, then recognition accuracy in challenging conditions improves, but processing time increases
Solution Approach 1:
The system performs preliminary action by first attempting facial recognition, which is faster and sufficient for many cases. Only when facial recognition fails or is unreliable does the system proceed to the more time-consuming bodyprint analysis. This sequential approach with early exit optimizes processing time while maintaining high accuracy in challenging conditions.
Solution Approach 2:
The system applies partial action by performing only facial analysis when it suffices (reducing processing time) and excessive action by adding bodyprint analysis only when necessary (improving accuracy). This conditional execution strategy resolves the contradiction between processing time and accuracy by avoiding unnecessary computational overhead.
Data Source
AI summary
Techniques are disclosed for determining whether to include a bodyprint in a cluster of bodyprints associated with a recognized person. For example, a device performs facial recognition to identify the identity of a first person. The device also identifies and stores physical characteristic information of the first person, the stored information associated with the identity of the first person based on the recognized face. Subsequently, the device receives a second video feed showing an image of a second person whose face is also determined to be recognized by the device. The device then generates a quality score for physical characteristics in the image of the user. The device can then add the image with the physical characteristics to a cluster of images associated with the person if the quality score is above a threshold, or discard the image if not.