Convolutional neural networks and training strategies for skin detection

Kim, Yoon-Sik; Hwang, Insung; Cho, Nam Ik

doi:10.1109/icip.2017.8297017

Cited by 27 publications

(22 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The utilization of 1 × 1 convolution leads to fewer parameters than other general CNNs with the same depth. This block has been successfully applied in different image restoration and labelling tasks, such as compression, artefacts reduction, semantic segmentation and skin detection [KHC17a,KHC17b]. Our network pipeline is depicted in Figure 4 and consists of two convolutional layers followed by eight inception blocks and a final convolutional layer.…”

Section: Calibrationmentioning

confidence: 99%

Transfer Deep Learning for Reconfigurable Snapshot HDR Imaging Using Coded Masks

Alghamdi

Thabet

et al. 2021

Computer Graphics Forum

View full text Add to dashboard Cite

High dynamic range (HDR) image acquisition from a single image capture, also known as snapshot HDR imaging, is challenging because the bit depths of camera sensors are far from sufficient to cover the full dynamic range of the scene. Existing HDR techniques focus either on algorithmic reconstruction or hardware modification to extend the dynamic range. In this paper we propose a joint design for snapshot HDR imaging by devising a spatially varying modulation mask in the hardware and building a deep learning algorithm to reconstruct the HDR image. We leverage transfer learning to overcome the lack of sufficiently large HDR datasets available. We show how transferring from a different large-scale task (image classification on ImageNet) leads to considerable improvements in HDR reconstruction. We achieve a reconfigurable HDR camera design that does not require custom sensors, and instead can be reconfigured between HDR and conventional mode with very simple calibration steps. We demonstrate that the proposed hardware-software so lution offers a flexible yet robust way to modulate per-pixel exposures, and the network requires little knowledge of the hardware to faithfully reconstruct the HDR image. Comparison results show that our method outperforms the state of the art in terms of visual perception quality.

show abstract

Section: Calibrationmentioning

confidence: 99%

Transfer Deep Learning for Reconfigurable Snapshot HDR Imaging Using Coded Masks

Alghamdi

Thabet

et al. 2021

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…Despite the presence of a huge number of pixel-based approaches (Kakumanu et al, 2007), the region-based skin color detection techniques are very few (Poudel, Zhang, Liu, & Nait-Charif, 2013)(W. C. Chen & Wang, 2007) (Kruppa, Bauer, & Schiele, 2002) (Sebe, Cohen, Huang, & Gevers, 2004). Some recent methods (Zuo et al, 2017) (Kim, Hwang, & Cho, 2017b) based on convolutional neural networks can be included in this category.…”

Section: Skin Detection Approachesmentioning

confidence: 99%

“…The first class of methods, based on sophisticated and computationally expensive techniques, include recently proposed approaches based on deep learning (Xu et al, 2015) (Zuo et al, 2017) (Kim et al, 2017b) (Ma & Shih, 2018). Convolutional neural networks have recently achieved remarkable results for a variety of computer vision tasks, including several applications based on pixel-wise prediction (i.e.…”

Section: Skin Detection Approachesmentioning

confidence: 99%

“…Anyway in none of the above surveys is there a fair comparison among methods using the same testing protocols and datasets. The aim of the present work is not limited to survey the most recent research in this field (which is now enriched of methods based on deep learning (Xu, Zhang, & Wang, 2015) (Zuo, Fan, Blasch, & Ling, 2017) (Kim, Hwang, & Cho, 2017a) (Ma & Shih, 2018)), but also, and above all, to propose a framework for a fair comparison among approaches.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fair comparison of skin detection approaches on publicly available datasets

Lumini

Nanni

2020

Expert Systems with Applications

View full text Add to dashboard Cite

“…The advantage of this combination was that the FCN layers captured local features while the RNN layers modeled the semantic contextual dependencies. In reference [ 12 ], an inception-based architecture was proposed that was composed of convolutional and inception modules with training considering both patches and whole images. In reference [ 13 ], the authors performed experiments using several CNN architectures and concluded that DeepLabv3+ is the best CNN for skin segmentation.…”

Section: Introductionmentioning

confidence: 99%

Postprocessing for Skin Detection

et al. 2021

View full text Add to dashboard Cite

Skin detectors play a crucial role in many applications: face localization, person tracking, objectionable content screening, etc. Skin detection is a complicated process that involves not only the development of apposite classifiers but also many ancillary methods, including techniques for data preprocessing and postprocessing. In this paper, a new postprocessing method is described that learns to select whether an image needs the application of various morphological sequences or a homogeneity function. The type of postprocessing method selected is learned based on categorizing the image into one of eleven predetermined classes. The novel postprocessing method presented here is evaluated on ten datasets recommended for fair comparisons that represent many skin detection applications. The results show that the new approach enhances the performance of the base classifiers and previous works based only on learning the most appropriate morphological sequences.

show abstract

Convolutional neural networks and training strategies for skin detection

Cited by 27 publications

References 25 publications

Transfer Deep Learning for Reconfigurable Snapshot HDR Imaging Using Coded Masks

Transfer Deep Learning for Reconfigurable Snapshot HDR Imaging Using Coded Masks

Fair comparison of skin detection approaches on publicly available datasets

Postprocessing for Skin Detection

Contact Info

Product

Resources

About