NTIRE 2021 Challenge on High Dynamic Range Imaging: Dataset, Methods and Results

Pérez-Pellitero, Eduardo; Catley-Chandar, Sibi; Leonardis, Aleš; Timofte, Radu; Wang, Xian; Li, Yong; Wang, Tao; Song, Fenglong; Liu, Zhen; Lin, Wenjie; Li, Xinpeng; Rao, Qing; Jiang, Tuo; Han, Mingyan; Fan, Haoqiang; Sun, Jian; Liu, Shuaicheng; Chen, Xiangyu; Liu, Yihao; Zhang, Zhengwen; Qiao, Yu; Dong, Chao; Chee, Evelyn Yi Lyn; Shen, Shanlan; Duan, Yubo; Chen, Guannan; Sun, Mengdi; Gao, Yan; Zhang, Lijie; Akhil, K A; Jiji, C. V.; Sharif, S. M. A.; Naqvi, Rizwan Ali; Biswas, Mithun; Kim, Sungjun; Xia, Chenjie; Zhao, Bowen; Ye, Zhangyu; Lu, Xiwen; Cao, Yanan; Yang, Jiangxin; S, Green Rosh K; Lomte, Sachin Deepak; Krishnan, Nikhil; Prasad, B. V. S. S. S.

doi:10.1109/cvprw53098.2021.00078

Cited by 50 publications

(35 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We participated in the NTIRE2021 HDR Challenge [40] and won the second place in the single frame track. The results are shown in Table 6.…”

Section: Results Of Ntire2021 Hdr Challengementioning

confidence: 99%

“…Compared to the other commonly used losses of image restoration, this loss can lead to better quantitative performance and visual quality. • Experiments show that our method outperforms the state-of-the-art methods both quantitatively and qualitatively, and we won the second place in the single frame track of NTIRE2021 HDR Challenge [40].…”

Section: Introductionmentioning

confidence: 86%

“…Previous studies [14,13,34,27] have adopted different datasets on the task of image HDR reconstruction for training and evaluation. In this paper, we use the dataset proposed by NITRE 2021 HDR Challenge [40]. As depicted in this challenge, the dataset is a subset of images selected from the HdM HDR dataset [15], where the HDR images are captured by two Alexa Arri cameras with a mirror rig and the corresponding LDR images are generated by applying a degradation model (e.g., exposure gain, noise addition and quantization, clipping).…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization

Chen¹,

Liu²,

Zhang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Figure 1. HDR reconstruction with denoising and dequantization from a single LDR image. We propose a novel learning based method for single image HDR reconstruction with denoising and dequantization. The proposed method consists of a spatially dynamic encoder-decoder network and a new T anh L1 loss function. The visual comparison shows that our method reconstructs information in over-exposed regions and also reduces the noise and quantization loss in well-exposed regions. All the images have been µ-law tone-mapped for display. We slightly increase the contrast of patches in the bottom row for clearer visualization. Please zoom in for best view.

show abstract

“…We participated in the NTIRE2021 HDR Challenge [40] and won the second place in the single frame track. The results are shown in Table 6.…”

Section: Results Of Ntire2021 Hdr Challengementioning

confidence: 99%

Section: Introductionmentioning

confidence: 86%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization

Chen¹,

Liu²,

Zhang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We train and evaluate our method on the dataset provided by the NTIRE2021 Multi-Frame HDR Challenge [27]. It contains 1463 valid scenes in total as we exclude 31 incomplete scenes.…”

Section: Dataset and Implementation Detailsmentioning

confidence: 99%

ADNet: Attention-guided Deformable Convolutional Network for High Dynamic Range Imaging

Liu¹,

Lin²,

Li³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we present an attention-guided deformable convolutional network for hand-held multi-frame high dynamic range (HDR) imaging, namely ADNet. This problem comprises two intractable challenges of how to handle saturation and noise properly and how to tackle misalignments caused by object motion or camera jittering. To address the former, we adopt a spatial attention module to adaptively select the most appropriate regions of various exposure low dynamic range (LDR) images for fusion. For the latter one, we propose to align the gamma-corrected images in the feature-level with a Pyramid, Cascading and Deformable (PCD) alignment module. The proposed AD-Net shows state-of-the-art performance compared with previous methods, achieving a PSNR-l of 39.4471 and a PSNRµ of 37.6359 in NTIRE 2021 Multi-Frame HDR Challenge. Recently, several learning-based methods have been explored. Kalantari et al. proposed the first deep convolu-

show abstract

“…This challenge is one of the NTIRE 2021 associated challenges: nonhomogeneous dehazing [3], defocus deblurring using dual-pixel [1], depth guided image relighting [15], image deblurring [36], multi-modal aerial view imagery classification [31], learning the super-resolution space [32], quality enhancement of heavily compressed videos [56], video super-resolution [49], perceptual image quality assessment [21], burst super-resolution [5], high dynamic range [37].…”

Section: Introductionmentioning

confidence: 99%

NTIRE 2021 Challenge on Perceptual Image Quality Assessment

Gu¹,

Cai²,

Chen³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper reports on the NTIRE 2021 challenge on perceptual image quality assessment (IQA), held in conjunction with the New Trends in Image Restoration and Enhancement workshop (NTIRE) workshop at CVPR 2021. As a new type of image processing technology, perceptual image processing algorithms based on Generative Adversarial Networks (GAN) have produced images with more realistic textures. These output images have completely different characteristics from traditional distortions, thus pose a new challenge for IQA methods to evaluate their visual quality. In comparison with previous IQA challenges, the training and testing datasets in this challenge include the outputs of perceptual image processing algorithms and the corresponding subjective scores. Thus they can be used to develop and evaluate IQA methods on GAN-based distortions. The challenge has 270 registered participants in total. In the final testing stage, 13 participating teams submitted their models and fact sheets. Almost all of them have achieved much better results than existing IQA methods, while the winning method can demonstrate state-of-the-art performance.

show abstract

NTIRE 2021 Challenge on High Dynamic Range Imaging: Dataset, Methods and Results

Cited by 50 publications

References 55 publications

HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization

HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization

ADNet: Attention-guided Deformable Convolutional Network for High Dynamic Range Imaging

NTIRE 2021 Challenge on Perceptual Image Quality Assessment

Contact Info

Product

Resources

About