Tracking of Fingertips and Centres of Palm using KINECT (1304.4662v1)

Abstract: Hand Gesture is a popular way to interact or control machines and it has been implemented in many applications. The geometry of hand is such that it is hard to construct in virtual environment and control the joints but the functionality and DOF encourage researchers to make a hand like instrument. This paper presents a novel method for fingertips detection and centres of palms detection distinctly for both hands using MS KINECT in 3D from the input image. KINECT facilitates us by providing the depth information of foreground objects. The hands were segmented using the depth vector and centres of palms were detected using distance transformation on inverse image. This result would be used to feed the inputs to the robotic hands to emulate human hands operation.

• The paper proposes a novel markerless method using Kinect's depth sensor to accurately track fingertips and palm centers for natural hand gesture recognition.
• The methodology involves depth acquisition, threshold-based segmentation, fingertip detection by minimum depth, and palm center identification via distance transform.
• The system achieves high accuracy, near-perfect for extended fingertips and ~90% for palm centers, showing potential for real-time robotic control and teleoperation.

Analysis of Fingertip and Palm Center Tracking Using Kinect

The paper "Tracking of Fingertips and Centre of Palm using KINECT" by Raheja, Chaudhary, and Singal addresses the problem of detecting dynamic hand gestures through a novel approach leveraging the Microsoft Kinect. The authors propose a method that capitalizes on the Kinect’s depth-sensing capabilities to facilitate the segmentation of hands, thereby identifying fingertips and centers of the palms in three-dimensional space without the need for sensors or markers. This advancement is presented under the concept of "Natural Computing," aiming for interaction with technology through unencumbered, natural hand movements.

The methodology introduced by the authors represents an evolution from conventional hand detection systems that typically suffer limitations under varied conditions such as quick hand motion, cluttered backgrounds, or insufficient lighting. Notably, the proposed model does not assume specific hand orientations, rendering it versatile compared to previous efforts which often relied on assumptions regarding hand positioning.

Methodology

The paper delineates a multi-step approach to achieve the desired detection:

1. Depth Acquisition: Utilizes Kinect's depth-sensing infrastructures, comprising an infrared camera and PrimeSense sensor, to acquire 3D depth images.
2. Segmentation: Employs depth slicing techniques, defining thresholds to isolate hands from the background.
3. Fingertip Detection: Identifies fingertips based on minimum depth values across detected fingers, using the depth information to deduce proximity to the camera.
4. Palm Center Detection: Applies a distance transform on inverted binary images to determine palm centers, ensuring clear differentiation for dual-hand scenarios by employing color coding for distinction.

Experimental Results

The implemented system demonstrates robust real-time detection, achieving impressive accuracy rates. Fingertip detection accuracy approaches near-perfect levels when fingers are fully extended, while palm center localization maintains approximately 90% fidelity. Results remain satisfactory despite fingers being bent at significant angles, emphasizing the system’s reliability.

Implications and Future Directions

The systems' high accuracy and real-time performance underscore its potential application in controlling robotic appendages through gestural inputs. Specifically, this might revolutionize manipulation tasks where human safety is a concern, such as in hazardous environments or intricate surgical procedures. Moreover, the paper hints at an ongoing project to regulate robotic hands purely through hand gestures, suggesting significant implications for both teleoperation and augmented reality interfaces.

Looking forward, future research could focus on refining the algorithm to handle more extreme environmental conditions or integrating additional features like gesture classification based on the tracked data. Additionally, exploring the scalability of this approach to capture gestures more comprehensively in diverse demographics and of varied cultural gestures could broaden its applicability.

In summary, the work presented by Raheja et al. extends the field of intuitive human-computer interaction by enhancing the capture and interpretation of natural hand gestures, paving the way for more immersive and natural interactions with technological systems.