Boosting performance of face detection by using an efficient skin segmentation algorithm

Mahmoodi, Mohammad Reza; Sayedi, Sayed Masoud

doi:10.1109/iciteed.2014.7007899

Cited by 12 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to corroborate the validity of the system, two image processing algorithms namely a skin detection and a motion detection is designed. For the former, the aim is to group all the pixels into two classes of skin and non-skin pixels for any image [26,27]. This has numerous applications in surveillance, content based coding, and face detection [28,29,30], etc.…”

Section: Experimental Setup and Resultsmentioning

confidence: 99%

A Low Cost High Speed FPGA-Based Image Processing Framework

Mahmoodi

Sayedi

2016

IJMECS

View full text Add to dashboard Cite

Abstract-In this paper, a high-speed and low-cost image processing framework based on MATLAB-FPGA interface is proposed that can be used in researches aiming at developing wide variety of not only image processing tasks but also many signal processing applications. In addition, this new framework could be exploited for several other tasks such as on-chip verification, using PC as an enormous external RAM for FPGA while preserving high speed data access, developing hardware-software co-designs, etc. The communication between FPGA and MATLAB is via 1Gbs Ethernet based on UDP/IP protocol which is very promising for high speed data transmission in point-topoint communications. UDP stack is efficiently designed in FPGA based on a fully pipelined architecture with minimum level of logic in order to reach high performance.. Dynamic data transmission between the UDP stack, memory and an arbitrary image processing module makes it possible to practically simulate, debug and implement most relevant applications. The hardware system is relatively low-cost and it consumes a negligible area of a Spartan-6 LXT45 Xilinx FPGA. Operating at 1 Gb/s, theoretically, the system is capable of processing 132 frames of 640*480 color images in a second. The effectiveness of the system is evaluated by means of both place and route simulation and practical implementation of a skin detection algorithm and a motion detector.

show abstract

Section: Experimental Setup and Resultsmentioning

confidence: 99%

A Low Cost High Speed FPGA-Based Image Processing Framework

Mahmoodi

Sayedi

2016

IJMECS

View full text Add to dashboard Cite

show abstract

“…This process is repeated for new adjacent pixels until no neighbouring pixels satisfy the conditions and the growth of the region stops. Variations of this method have been proposed in [34][35][36] where different measurement techniques have been proposed to measure the similarity between adjacent pixels for assigning them to a region, such as Euclidean distance, colour distance map or probability measuring.…”

Section: Related Workmentioning

confidence: 99%

Weakly Supervised Human Skin Segmentation Using Guidance Attention Mechanisms

Kooshan

Climent-Pérez

Flórez-Revuelta

2023

Preprint

View full text Add to dashboard Cite

“…This process is repeated for new adjacent pixels until no neighboring pixels satisfy the conditions and the growth of the region stops. Variations of this method have been proposed in [34,35,36] where different measurement techniques have been proposed to measure the similarity between adjacent pixels for assigning them to a region, such as Euclidean distance, color distance map or probability measuring.…”

Section: Related Workmentioning

confidence: 99%

Weakly Supervised Human Skin Segmentation using Guidance Attention Mechanisms

Kooshan¹,

Climent-Pérez²,

Flórez-Revuelta³

2023

Preprint

View full text Add to dashboard Cite

Human skin segmentation is a crucial task in computer vision and biometric systems, yet it poses several challenges such as variability in skin color, pose, and illumination. This paper presents a robust data-driven skin segmentation method for a single image that addresses these challenges through the integration of contextual information and efficient network design. In addition to robustness and accuracy, the integration into real-time systems requires a careful balance between computational power, speed, and performance. The proposed method incorporates two attention modules, Body Attention and Skin Attention, that utilize contextual information to improve segmentation results. These modules draw attention to the desired areas, focusing on the body boundaries and skin pixels, respectively. Additionally, an efficient network architecture is employed in the encoder part to minimize computational power while retaining high performance. To handle the issue of noisy labels in skin datasets, the proposed method uses a weakly supervised training strategy, relying on the Skin Attention module. The results of this study demonstrate that the proposed method is comparable to, or outperforms, state-of-the-art methods on benchmark datasets.

show abstract

Boosting performance of face detection by using an efficient skin segmentation algorithm

Cited by 12 publications

References 44 publications

A Low Cost High Speed FPGA-Based Image Processing Framework

A Low Cost High Speed FPGA-Based Image Processing Framework

Weakly Supervised Human Skin Segmentation Using Guidance Attention Mechanisms

Weakly Supervised Human Skin Segmentation using Guidance Attention Mechanisms

Contact Info

Product

Resources

About