Hierarchical Density-Aware Dehazing Network

Zhang, George; Ren, Wenqi; Zhang, Shengdong; Zhang, He; Nie, Yunfeng; Xue, Zhe; Cao, Xiaochun

doi:10.1109/tcyb.2021.3070310

Cited by 31 publications

(21 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For comparison, after the warranty connection, the upper, lower, and all parts of the body were selected to exclude RGB functions, SIFT functions, flow features, and SIFT and optical flow features. By recognizing the value obtained by various methods, it can be seen that the order is close to being received after the upper body or lower body is used for special mining, but the body acceptance is slightly lower [22]. e combination of human body parts has the highest cognitive value, while the combination of human body parts has the lowest cognitive value.…”

Section: Experimental Results and Analysismentioning

confidence: 87%

A Hierarchical Children’s Dance Movement Pose Estimation Method Based on Sequence Multiscale Feature Fusion Representation

Qin

Tao

Tang

2022

Advances in Multimedia

View full text Add to dashboard Cite

Aiming at the problem that traditional human motion pose estimation methods cannot accurately capture and estimate the movement changes of children dancers, a hierarchical dance pose estimation method for children based on sequence multiscale feature fusion representation is proposed. By comparing the pose feature extraction algorithm with the actual recognition effect, the recognition rates of the dancer’s upper body, infiltration, and whole body have increased by 14.2, 10.6, and 12.6, respectively. The experimental results show that the proposed pose estimation algorithm achieves good pose estimation results on both the standard human pose estimation dataset and the self-built dance dataset.

show abstract

Section: Experimental Results and Analysismentioning

confidence: 87%

A Hierarchical Children’s Dance Movement Pose Estimation Method Based on Sequence Multiscale Feature Fusion Representation

Qin

Tao

Tang

2022

Advances in Multimedia

View full text Add to dashboard Cite

show abstract

“…In [47], the authors presented a Dual-Path Recurrent network, which has two parallel branches to recover the image by simultaneously learning the characteristics of the basic content and details of foggy images. Zhang et al [45] proposed a densely connected pyramid encoder to extract multiscale image features from foggy inputs before using a density-aware generator to obtain fog-free images; they also constructed a Laplacian pyramid decoder to refine the defogged images. Different from the above methods, Qu et al [34] treated image defogging as an image-to-image translation problem and proposed an enhanced pix2pix defogging network.…”

Section: B Supervised Methodsmentioning

confidence: 99%

“…It is thus necessary to design effective single image defogging algorithms to restore the content, color, and texture details from the foggy images. Existing image defogging methods can be roughly divided into three classes: prior-based methods [15], [4], [54], [19], [2], fusion-based methods [1], [10], [11], and learning-based methods [5], [37], [9], [52], [45], [47]. Most prior-based methods remove the fog from an image by using or improving an atmospheric scattering model (ASM) [29].…”

Section: Introductionmentioning

confidence: 99%

“…These methods can be grouped into two categories: supervised and unsupervised defogging methods: for former requires pairs of fog and fog-free images in network training, whereas the latter does not. In addition, supervised defogging mainly includes ASM based defogging networks (ASMDN) [5] [36] [52] [21] and end-to-end defogging networks (EDN) [37] [41] [34] [47] [45]. In ASM, transmission and atmospheric light are two most important parameters.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Holistic Attention-Fusion Adversarial Network for Single Image Defogging

Liu¹,

Chen²,

Jiang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Adversarial learning-based image defogging methods have been extensively studied in computer vision due to their remarkable performance. However, most existing methods have limited defogging capabilities for real cases because they are trained on the paired clear and synthesized foggy images of the same scenes. In addition, they have limitations in preserving vivid color and rich textual details in defogging. To address these issues, we develop a novel generative adversarial network, called holistic attention-fusion adversarial network (HAAN), for single image defogging. HAAN consists of a Fog2Fogfree block and a Fogfree2Fog block. In each block, there are three learningbased modules, namely, fog removal, color-texture recovery, and fog synthetic, that are constrained each other to generate high quality images. HAAN is designed to exploit the self-similarity of texture and structure information by learning the holistic channel-spatial feature correlations between the foggy image with its several derived images. Moreover, in the fog synthetic module, we utilize the atmospheric scattering model to guide it to improve the generative quality by focusing on an atmospheric light optimization with a novel sky segmentation network. Extensive experiments on both synthetic and real-world datasets show that HAAN outperforms state-of-the-art defogging methods in terms of quantitative accuracy and subjective visual quality.

show abstract

“…Recently, most image dehazing works are reported remarkable results on synthesized hazy images [6][7][8][9][10][11][12][13][14][15][16][17][18][19], but almost no one make objective evaluations on real-world hazy images. In this work, the RW-Haze dataset is exploited for a comprehensive dehazing assessment.…”

Section: Evaluated Dehazing Mehtodsmentioning

confidence: 99%

RW-HAZE: A Real-World Benchmark Dataset to Evaluate Quantitatively Dehazing Algorithms

Chen

Wang

Liu

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Most existing image dehazing approaches are able to achieve desirable results whose differences are too subtle for people to qualitatively judge. Therefore, it is important to adopt quantitative assessment on real-world hazy images. However, many dehazing works have not been quantitatively evaluated on real-world hazy images due to the lack of appropriate realworld datasets. In this work, we attempt to address the issue and present a well-aligned real-world benchmark dataset, namely RW-Haze, for image dehazing evaluation, which had been lacking for a long period of time. It contains 210 pairs of well-aligned haze-free images and hazy images with distinct haze densities, which were captured from six cities in China by fixed cameras. To the best of our knowledge, RW-Haze is the first real-world dataset that is made up of well-aligned image pairs of haze-free and hazy images with diverse haze levels. We select 13 state-of-the-art single image dehazing works for making comprehensive evaluations among them on RW-Haze dataset. Experimental results show that there still exist rich rooms for image dehazing research to improve its robustness on natural hazy images, especially dense haze scenes.

show abstract

Hierarchical Density-Aware Dehazing Network

Cited by 31 publications

References 57 publications

A Hierarchical Children’s Dance Movement Pose Estimation Method Based on Sequence Multiscale Feature Fusion Representation

A Hierarchical Children’s Dance Movement Pose Estimation Method Based on Sequence Multiscale Feature Fusion Representation

Holistic Attention-Fusion Adversarial Network for Single Image Defogging

RW-HAZE: A Real-World Benchmark Dataset to Evaluate Quantitatively Dehazing Algorithms

Contact Info

Product

Resources

About