Labeled from Unlabeled: Exploiting Unlabeled Data for Few-shot Deep HDR Deghosting

Prabhakar, K. Ram; Senthil, Gowtham; Agrawal, Susmit; Babu, R. Venkatesh; Gorthi, Rama Krishna Sai Subrahmanyam

doi:10.1109/cvpr46437.2021.00484

Cited by 15 publications

(8 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compare our results with previous state-of-the-art methods, including two patch-based methods [42,15] and seven CNN-based approaches [18,45,47,49,36,34,38]. Note that Kalantari et al [18] and Prabhakar et al [36] first align the input images using optical flow and Wu et al [45] apply homography transformation.…”

Section: Comparisonsmentioning

confidence: 99%

“…Note that Kalantari et al [18] and Prabhakar et al [36] first align the input images using optical flow and Wu et al [45] apply homography transformation. Also, Prabhakar et al [38] use both of them for pre-alignment. We used the official codes if they are provided.…”

Section: Comparisonsmentioning

confidence: 99%

“…Since the direct acquisition of HDR images is practically tricky and requires expensive imaging devices, HDR imaging techniques using low dynamic range (LDR) images are drawing considerable attention. There have been many methods to generate an HDR image from a single LDR input for this reason, where earlier methods just stretched the dynamic range of the LDR input [24,1,31,10,5,6,41,16,22,23], and some recent methods (a) Input LDR images (b) Yan [47] (c) Prabhakar [38] (d) Ours…”

Section: Introductionmentioning

confidence: 99%

“…Wu et al [45] adopted homography transformation for background alignment and the U-Net for HDR reconstruction. Prabhakar et al [38] used both homography and optical flow. Meanwhile, Yan et al [47,49] utilized attention modules and non-local blocks, respectively, to implicitly align the input features.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

High Dynamic Range Imaging of Dynamic Scenes with Saturation Compensation but without Explicit Motion Compensation

Chung

Cho

2022

2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

High dynamic range (HDR) imaging is a highly challenging task since a large amount of information is lost due to the limitations of camera sensors. For HDR imaging, some methods capture multiple low dynamic range (LDR) images with altering exposures to aggregate more information. However, these approaches introduce ghosting artifacts when significant inter-frame motions are present. Moreover, although multi-exposure images are given, we have little information in severely over-exposed areas. Most existing methods focus on motion compensation, i.e., alignment of multiple LDR shots to reduce the ghosting artifacts, but they still produce unsatisfying results. These methods also rather overlook the need to restore the saturated areas. In this paper, we generate well-aligned multi-exposure features by reformulating a motion alignment problem into a simple brightness adjustment problem. In addition, we propose a coarse-to-fine merging strategy with explicit saturation compensation. The saturated areas are reconstructed with similar well-exposed content using adaptive contextual attention. We demonstrate that our method outperforms the state-of-the-art methods regarding qualitative and quantitative evaluations.

show abstract

Section: Comparisonsmentioning

confidence: 99%

Section: Comparisonsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

High Dynamic Range Imaging of Dynamic Scenes with Saturation Compensation but without Explicit Motion Compensation

Chung

Cho

2022

2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

show abstract

“…The methods in the first group [19,23,24,26,29,30,31,32,33,34,35,36,37,38] suffer from the problem of not accurately processing over-and under-exposed regions, but provide more compact systems. Methods in the second group use a group of bracketed overand under-exposed images as input to directly learn to generate HDR output [39,39,40,41,42,43,44,45,46], as similarly utilized in photographic HDR generation. Methods belonging to these two groups differ mainly according to their network components, such as non-local blocks [42] and attention mechanisms [40].…”

Section: Related Workmentioning

confidence: 99%

A Mixed Quantization Network for Computationally Efficient Mobile Inverse Tone Mapping

Borrego-Carazo¹,

Özay²,

Laboyrie³

et al. 2022

Preprint

View full text Add to dashboard Cite

Recovering a high dynamic range (HDR) image from a single low dynamic range (LDR) image, namely inverse tone mapping (ITM), is challenging due to the lack of information in over-and under-exposed regions. Current methods focus exclusively on training high performing but computationally inefficient ITM models, which in turn hinder deployment of the ITM models in resource-constrained environments with limited computing power such as edge and mobile device applications.To this end, we propose combining efficient operations of deep neural networks with a novel mixed quantization scheme to construct a well performing but computationally efficient mixed quantization network (MQN) which can perform single image ITM on mobile platforms. In the ablation studies, we explore the effect of using different attention mechanisms, quantization schemes and loss functions on the performance of MQN in ITM tasks. In the comparative analyses, ITM models trained using MQN perform on par with the state-of-the-art methods on benchmark datasets. MQN models provide up to 10 times improvement on latency and 25 times improvement on memory consumption.

show abstract

Ghost-free High Dynamic Range Imaging with Context-Aware Transformer

Liu¹,

Wang

Zeng

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Labeled from Unlabeled: Exploiting Unlabeled Data for Few-shot Deep HDR Deghosting

Cited by 15 publications

References 36 publications

High Dynamic Range Imaging of Dynamic Scenes with Saturation Compensation but without Explicit Motion Compensation

High Dynamic Range Imaging of Dynamic Scenes with Saturation Compensation but without Explicit Motion Compensation

A Mixed Quantization Network for Computationally Efficient Mobile Inverse Tone Mapping

Ghost-free High Dynamic Range Imaging with Context-Aware Transformer

Contact Info

Product

Resources

About