Non-local Image Dehazing

By Dana Berman, Tali Treibitz, Shai Avidan. Computer Vision and Pattern Recognition, 2016

1,515 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7b2ca78221fc59b40c122e3b230b8f552e856d12

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

FFA-Net: Feature Fusion Attention Network for Single Image Dehazing (2019, background): “…the optical transmission in hazy scenes, the scattered light is eliminated to increase scene visibility and recover haze-free scene contrasts., (Berman, Avidan, and others 2016) proposed a non-local prior to characterize the clean image, the algorithm relies on the assumption that colors of a…”
MAT: Mask-Aware Transformer for Large Hole Image Inpainting (2022, background): “Creating reasonable structures and textures for the missing areas demands contextual understanding, using distant information according to non-local priors [4, 7, 38, 56] in images.”
Effects of Image Degradations to CNN-based Image Classification (2018, methodology): “As discussed earlier, for many kinds of image degradations, many researches have been conducted to remove/reduce the degradation to restore the underlying clear images [15], [16], [17], [18], [19], [20], [21], [22], [25], [37], [38], [40], [42], [44], [45].”
Benchmarking Single-Image Dehazing and Beyond (2017, methodology): “DCP [9] FVR [10] BCCR [11] GRM [12] CAP [13] NLD [14] DehazeNet [15] MSCNN [16] AOD-Net [17] PSNR 16.”
Flash Photography for Data-Driven Hidden Scene Recovery (2018, background): “NLOS imaging has been explored widely in the context of vision related applications such as: image dehazing [1, 5, 9, 32], seeing through translucent objects [10], cloud tomography [16, 27], underwater imaging [33], and tracking [12, 19].”
Gated Fusion Network for Single Image Dehazing (2018, methodology): “Since we directly restore the ﬁnal dehazed image without transmission estimation in our algorithm, we only compare the ﬁnal dehazed results with other methods. The priors based image dehazing methods [10,21,3] overestimate the haze thickness, so the dehazed results tend to be darker than the ground ”
Image Dehazing Transformer with Transmission-Aware 3D Position Embedding (2022, methodology): “We only adopt the three-stage Swin Transformer, where the lightweight transformer parameters are adopted, i.e., the depth and numbers of attention heads are set to [2,2,2] and [3,6,9], respectively.”
You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network (2020, methodology): “, 2016), NLD (Berman et al., 2016).”
NH-HAZE: An Image Dehazing Benchmark with Non-Homogeneous Hazy and Haze-Free Images (2020, methodology): “[13] solution is based on the color consistency observation introduced by Omer et al .”
I-HAZE: a dehazing benchmark with real hazy and haze-free indoor images (2018, methodology): “Berman et al. [20] introduced an algorithm that also use local airlight estimation, and extends the color consistency observation of Fattal [14].”
Underwater Image Enhancement With Hyper-Laplacian Reflectance Priors (2022, positive): “Then, we respectively use the SIFT keypoint detection [46] and the transmission estimation [48] to detect underwater keypoints of the enhanced images by different methods and estimate the corresponding depth maps.”
Domain Adaptation for Image Dehazing (2020, methodology): “From Figure 5 (b), we can observe that both NLD [2] and GFN [27] suffer from some color distortion where the result looks unrealistic.”
All in One Bad Weather Removal Using Architectural Search (2020, background): “This is also true for bad weather restoration or image enhancement, such as dehazing [47, 3, 1, 57, 7, 13, 26, 43], removal of raindrop and dirt [20, 18, 59, 8, 53, 30, 50, 62, 36, 4], of moderate rain [44, 29], and of heavy rain [25, 54].”
PSD: Principled Synthetic-to-Real Dehazing Guided by Physical Priors (2021, background): “References [1] Dana Berman, Tali Treibitz, and Shai Avidan.”
AOD-Net: All-in-One Dehazing Network (2017, methodology): “We compared the proposed model with several stateof-the-art dehazing methods: Fast Visibility Restoration (FVR) [25], Dark-Channel Prior (DCP) [8], Boundary Constrained Context Regularization (BCCR) [12], Automatic Atmospheric Light Recovery (ATM) [22], Color Attenuation Prior (CAP) [32], Non-local ”
Fast End-to-End Trainable Guided Filter (2018, methodology): “DGF takes a Method L 0 smoothing [40] Detail Manipulation [13] Style Transfer [3] Non-local Dehazing [5] Image similar amount of memory space to DBL but grows slower as the resolution increases.”
DEA-Net: Single Image Dehazing Based on Detail-Enhanced Convolution and Content-Guided Attention (2023, background): “The widely known priors include dark channel prior (DCP) [19], [20], non-local prior (NLP) [21], color attenuation prior (CAP) [22], etc. Specifically, He et al. [19], [20] propose DCP, leveraging a crucial observation that most local patches in haze-free outdoor images contain pixels with very low…”
Dense-Haze: A Benchmark for Image Dehazing with Dense-Haze and Haze-Free Images (2019, methodology): “In Fig.2 are shown 7 scenes of the Dense-Haze dataset and the dehazed image results generated using the methods of He et al. [9], Meng et al. [13], Fattal [15], Cai et al. [20], Ancuti et al. [33], Berman et al. [16] and Ren et al. [21].”