NonLocal Channel Attention for NonHomogeneous Image Dehazing

Metwaly, Kareem; Li, Xuelu; Guo, Tiantong; Monga, Vishal

doi:10.1109/cvprw50498.2020.00234

Cited by 23 publications

(9 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1) and AMIDR-Net (Fig. 3) are similar in terms of the general architecture which is inspired by U-Net [41,31]. An encoder is shared by three bottlenecks followed by four decoders.…”

Section: Network Architecturementioning

confidence: 99%

“…Table 2 details the structure of decoders. Each decoder includes four levels of cascaded attention module (Squeeze and excitation [17] or dilation inception modules [30]), a dense transitional block, and two residual blocks. It means that there…”

Section: Network Architecturementioning

confidence: 99%

“…High dynamic range (HDR) imaging algorithms [32,23] focus on increasing the local contrast of a low dynamic range image. Dehazing algorithms [30,48] seek for removing the haze artifacts caused by floating particles in the atmosphere. Shadow removal [26] and light enhancement [15] methods focus on enhancing the lighting and removing the artifacts in the image with the existing light source.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Physically Inspired Dense Fusion Networks for Relighting

Yazdani¹,

Guo²,

Monga³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Image relighting has emerged as a problem of significant research interest inspired by augmented reality applications. Physics-based traditional methods, as well as black box deep learning models, have been developed. The existing deep networks have exploited training to achieve a new state of the art; however, they may perform poorly when training is limited or does not represent problem phenomenology, such as the addition or removal of dense shadows. We propose a model which enriches neural networks with physical insight. More precisely, our method generates the relighted image with new illumination settings via two different strategies and subsequently fuses them using a weight map (w). In the first strategy, our model predicts the material reflectance parameters (albedo) and illumination/geometry parameters of the scene (shading) for the relit image (we refer to this strategy as intrinsic image decomposition (IID)). The second strategy is solely based on the black box approach, where the model optimizes its weights based on the ground-truth images and the loss terms in the training stage and generates the relit output directly (we refer to this strategy as direct). While our proposed method applies to both one-to-one and any-to-any relighting problems, for each case we introduce problemspecific components that enrich the model performance: 1) For one-to-one relighting we incorporate normal vectors of the surfaces in the scene to adjust gloss and shadows accordingly in the image. 2) For any-to-any relighting, we propose an additional multiscale block to the architecture to enhance feature extraction. Experimental results on the VIDIT 2020 and the VIDIT 2021 dataset (used in the NTIRE 2021 relighting challenge) reveals that our proposal can outperform many state-of-the-art methods in terms of wellknown fidelity metrics and perceptual loss.

show abstract

“…1) and AMIDR-Net (Fig. 3) are similar in terms of the general architecture which is inspired by U-Net [41,31]. An encoder is shared by three bottlenecks followed by four decoders.…”

Section: Network Architecturementioning

confidence: 99%

Section: Network Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Physically Inspired Dense Fusion Networks for Relighting

Yazdani¹,

Guo²,

Monga³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The method is based on two proposed models following similar ideology [43], the 'AtJw' and the 'AtJwD' models. As illustrated in Fig.…”

Section: Ipal-nonlocalmentioning

confidence: 99%

NTIRE 2020 Challenge on NonHomogeneous Dehazing

Ancuti¹,

Vasluianu²,

Timofte³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

This paper reviews the NTIRE 2020 Challenge on Non-Homogeneous Dehazing of images (restoration of rich details in hazy image). We focus on the proposed solutions and their results evaluated on NH-Haze, a novel dataset consisting of 55 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. NH-Haze is the first realistic nonhomogeneous haze dataset that provides ground truth images. The nonhomogeneous haze has been produced using a professional haze generator that imitates the real conditions of haze scenes. 168 participants registered in the challenge and 27 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.

show abstract

“…Recently, Convolutional Neural Networks (CNN) have proven their capability in extracting better features that ease the following step of classification and detection. This has been empirically proven in various fields, such as in object classification [45,27], object detection [25,71] and inverse image problems such as dehazing [58,99], denoising [51,73], HDR estimation [52,59,11] ... etc. Having said that, learning techniques typically require a huge amount of data for training and/or regularization of the training such as [7,97,3,69,8].…”

Section: Introductionmentioning

confidence: 99%

CAR -- Cityscapes Attributes Recognition A Multi-category Attributes Dataset for Autonomous Vehicles

Metwaly¹,

Kim²,

Branson³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Self-driving vehicles are the future of transportation. With current advancements in this field, the world is getting closer to safe roads with almost zero probability of having accidents and eliminating human errors. However, there is still plenty of research and development necessary to reach a level of robustness. One important aspect is to understand a scene fully including all details. As some characteristics (attributes) of objects in a scene (drivers' behavior for instance) could be imperative for correct decision making. However, current algorithms suffer from low-quality datasets with such rich attributes. Therefore, in this paper, we present a new dataset for attributes recognition -Cityscapes Attributes Recognition (CAR). The new dataset extends the well-known dataset Cityscapes by adding an additional yet important annotation layer of attributes of objects in each image. Currently, we have annotated more than 32k instances of various categories (Vehicles, Pedestrians, etc.). The dataset has a structured and tailored taxonomy where each category has its own set of possible attributes. The tailored taxonomy focuses on attributes that is of most beneficent for developing better self-driving algorithms that depend on accurate computer vision and scene comprehension. We have also created an API for the dataset to ease the usage of CAR. The API can be accessed through https://github.com/kareem-metwaly/CAR-API * Kareem Metwaly was in an internship at Scale AI through this work.

show abstract

NonLocal Channel Attention for NonHomogeneous Image Dehazing

Cited by 23 publications

References 42 publications

Physically Inspired Dense Fusion Networks for Relighting

Physically Inspired Dense Fusion Networks for Relighting

NTIRE 2020 Challenge on NonHomogeneous Dehazing

CAR -- Cityscapes Attributes Recognition A Multi-category Attributes Dataset for Autonomous Vehicles

Contact Info

Product

Resources

About