Hand Gesture Recognition for IPTV Using Trajectory Buffer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gesture recognition systems for remote control of IPTV are complex, require special equipment, and struggle with varying lighting conditions, unintentional operations, and the need for visual feedback, making them unsuitable for low-grade equipment and natural bare-hand gestures.
Innovation Solution
A gesture recognition system that uses a trajectory buffer to store hand centroids, employs adaptive fusion of motion and color cues, and relies on statistical calculations and predefinitions to identify hand gestures without visual feedback, suitable for low-grade equipment and varying lighting conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional gesture recognition systems use special equipment such as gloves or fixed color backgrounds, then gesture detection accuracy is improved, but device complexity and ease of operation deteriorate
Solution Approach 1:
The patent extracts and removes the requirement for special equipment (gloves, colored backgrounds) from the gesture recognition system. Instead, it uses only a standard camera to capture hand gestures by detecting skin color tones and motion patterns, thereby maintaining measurement precision while eliminating device complexity
Solution Approach 2:
The system makes the camera serve multiple functions: it captures both color information for skin detection and motion information for gesture tracking. This universal approach eliminates the need for specialized equipment while maintaining gesture detection accuracy
2Measurement precision
If gesture recognition systems use tracking methods such as optical flow or particle filtering, then tracking accuracy is improved, but computational complexity and processing time worsen
Solution Approach 1:
The patent extracts only the essential motion information needed for gesture recognition, discarding the computationally intensive optical flow and particle filtering methods. It uses simplified motion detection based on frame differencing and centroid tracking, achieving real-time processing while maintaining sufficient tracking accuracy
Solution Approach 2:
Instead of performing complete and exhaustive motion analysis, the system performs partial action by detecting only the hand region and its centroid movement. This selective approach reduces computational complexity while maintaining the necessary tracking accuracy for gesture recognition
3Measurement precision
If gesture recognition systems use background subtraction methods, then object detection is improved, but reliability deteriorates due to false detection of other moving objects
Solution Approach 1:
The patent applies local quality by focusing detection only on the hand region identified through skin color segmentation. Instead of monitoring the entire scene for motion, it restricts motion analysis to the locally identified hand area, thereby eliminating false detections from other moving objects while maintaining detection accuracy
Solution Approach 2:
The system introduces skin color detection as an intermediary step between background subtraction and motion tracking. This intermediary filters the scene to isolate only hand regions, preventing false detection of other moving objects and improving gesture recognition reliability
4Ease of operation
If gesture recognition requires visual feedback such as cursor position or hand image display, then ease of operation is improved, but adaptability and ease of operation worsen due to active area design requirements
Solution Approach 1:
The patent extracts and removes the visual feedback mechanism (cursor, hand image display) from the system. It relies on direct gesture recognition and interpretation without requiring active areas or visual cues, thereby maintaining ease of operation through natural gestures while improving adaptability to different interfaces
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a gesture recognition system for remote controlling a TV, which comprises: (a) video capturing device for capturing a video sequence of a user which is composed of video frames, and conveying the same into a hand tracker; (b) hand tracker module for receiving said video sequence, inspecting each of said video frames for possible inclusion of a user hand within it, and when a hand is determined to be found in a frame, calculating the hand centroid, and outputting the centroid coordinates into a trajectory buffer; (c) trajectory buffer for receiving from said hand tracker each of said hand centroid coordinates, and maintaining in storage at least several tens of recent hand centroids, that altogether define a trajectory; (d) trajectory segmentation module for continuously inspecting said buffer for possible inclusion of relevant trajectories that are suspected to relate to a hand gesture pattern, and whenever a relevant trajectory is detected, extracting a trajectory segment of the same, and conveying it into a gesture classifier; and (e) gesture classifier for extracting from each received trajectory segment global feature parameters of the same, comparing said extracted global feature parameters with gesture global predefinitions, concluding whether the segment relates to a specific gesture command, and in the affirmative case, transferring the command to the TV.