Filters








1,840 Hits in 5.5 sec

Datasets and Benchmarks for Densely Sampled 4D Light Fields [article]

Sven Wanner, Stephan Meister, Bastian Goldluecke
2013 International Symposium on Vision, Modeling, and Visualization  
The data is characterised by a dense sampling of the light fields, which best fits current plenoptic cameras and is a characteristic property not found in current multi-view stereo benchmarks.  ...  We present a new benchmark database to compare and evaluate existing and upcoming algorithms which are tailored to light field processing.  ...  To alleviate the above shortcomings, we present a new benchmark database which consists at the moment of 13 high quality densely sampled light fields.  ... 
doi:10.2312/pe.vmv.vmv13.225-226 dblp:conf/vmv/WannerMG13 fatcat:blxxtlyf3fczflkxo5toa3ej4q

A Study of Efficient Light Field Subsampling and Reconstruction Strategies [article]

Yang Chen, Martin Alain, Aljosa Smolic
2020 arXiv   pre-print
In this paper, we study subsampling and reconstruction strategies for light fields.  ...  Limited angular resolution is one of the main obstacles for practical applications of light fields.  ...  As for the synthetic LF dataset, all 28 light fields from the HCI benchmark [Honauer et al., 2016] were used. 10 light fields in total were selected from these datasets as the test set and the rest as  ... 
arXiv:2008.04694v1 fatcat:broru4bffvhojc7bzbswdob7le

Depth-based refocusing for reducing directional aliasing artifacts

Ensun Lee, Seohee Yang, Miseon Han, Jeongtae Kim
2016 Optics Express  
Meister, and B. Goldluecke, “Datasets and benchmarks for densely sampled 4D light fields,” in Proceedings of Vision Modeling and Visualization (VMV, 2013), pp. 225–226. S. Wanner and B.  ...  Meister, and B. Goldluecke, “Datasets and benchmarks for densely sampled 4D light fields,” in Proceedings of Vision Modeling and Visualization (VMV, 2013), pp. 225–226. S. Wanner and B.  ... 
doi:10.1364/oe.24.028065 pmid:27906372 fatcat:we5zlqcjwje4hbdgz3jvq43w2q

Fast and Efficient Depth Map Estimation from Light Fields [article]

Yuriy Anisimov, Didier Stricker
2018 arXiv   pre-print
The paper presents an algorithm for depth map estimation from the light field images in relatively small amount of time, using only single thread on CPU.  ...  The proposed method improves existing principle of line fitting in 4-dimensional light field space. Line fitting is based on color values comparison using kernel density estimation.  ...  The authors are grateful to Vladislav Golyanik, Kiran Varanasi and Jonathan Wray for the provided help.  ... 
arXiv:1805.00264v1 fatcat:ork5dh2xtzdupmo75ld6wzxjka

Sparse to Dense Scene Flow Estimation From Light Fields

Pierre David, Mikael Le Pendu, Christine Guillemot
2019 2019 IEEE International Conference on Image Processing (ICIP)  
A dataset of synthetic video light fields created for assessing scene flow estimation techniques is also described.  ...  The paper addresses the problem of scene flow estimation from sparsely sampled video light fields.  ...  EVALUATION Scene Flow Dataset For our experiments, we have prepared a synthetic video Light Field dataset 1 .  ... 
doi:10.1109/icip.2019.8803520 dblp:conf/icip/DavidPG19 fatcat:wc6y2a24nrg2dgy7qmrwpkjmzu

Dense Light Field Reconstruction From Sparse Sampling Using Residual Network [article]

Mantang Guo, Hao Zhu, Guoqing Zhou, Qing Wang
2018 arXiv   pre-print
Besides, reconstructing a large amount of light rays equivalent to multiple light fields using sparse sampling arises a severe challenge for existing methods.  ...  A light field records numerous light rays from a real-world scene. However, capturing a dense light field by existing devices is a time-consuming process.  ...  highest among light field benchmark datasets so far.  ... 
arXiv:1806.05506v2 fatcat:ckdsvcr5l5efzf26bgmrdmt4si

Attention-Based View Selection Networks for Light-Field Disparity Estimation

Yu-Ju Tsai, Yu-Lun Liu, Ming Ouhyoung, Yung-Yu Chuang
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Experiments show that the proposed method achieves state-of-the-art performance in terms of accuracy and ranks the first on a popular benchmark for disparity estimation for light field images.  ...  This paper introduces a novel deep network for estimating depth maps from a light field image.  ...  Acknowledgements This work was supported in part by Ministry of Science and Technology (MOST) and MOST Joint Research Center for  ... 
doi:10.1609/aaai.v34i07.6888 fatcat:bkebmzfv3jbjrhzh2d4aay3fku

Flexible Light Field Angular Superresolution via a Deep Coarse-to-Fine Framework

Qian Wang, Li Fang, Long Ye, Wei Zhong, Fei Hu, Qin Zhang, Ivan Lee
2022 Wireless Communications and Mobile Computing  
Acquisition of densely-sampled light fields (LFs) is challenging.  ...  In this paper, we develop a coarse-to-fine light field angular superresolution that reconstructs densely-sampled LFs from sparsely-sampled ones.  ...  SQ2020YFF0426386), and the Fundamental Research Funds for the Central Universities (Grant Nos. CUC19ZD006 and CUC21GZ007).  ... 
doi:10.1155/2022/4570755 fatcat:k3ijs4abkrgfthshjfjjms7rz4

Self-supervised Light Field View Synthesis Using Cycle Consistency [article]

Yang Chen, Martin Alain, Aljosa Smolic
2020 arXiv   pre-print
However, collecting such large datasets for light fields is challenging compared to natural images or videos.  ...  The proposed method aims to transfer prior knowledge learned from high quality natural video datasets to the light field view synthesis task, which reduces the need for labeled light field data.  ...  For this purpose, we use a variety of real-world and synthetic dense light field datasets which we sub-sample to create our test sparse datasets with sampling ratios α = 2 and α = 4.  ... 
arXiv:2008.05084v1 fatcat:4vyotfvkfbbp5oao5m533sv2h4

Self-supervised Light Field View Synthesis Using Cycle Consistency

Yang Chen, Martin Alain, Aljosa Smolic
2020 2020 IEEE 22nd International Workshop on Multimedia Signal Processing (MMSP)  
However, collecting such large datasets for light fields is challenging compared to natural images or videos.  ...  The proposed method aims to transfer prior knowledge learned from high quality natural video datasets to the light field view synthesis task, which reduces the need for labeled light field data.  ...  For this purpose, we use a variety of real-world and synthetic dense light field datasets which we sub-sample to create our test sparse datasets with sampling ratios α = 2 and α = 4.  ... 
doi:10.1109/mmsp48831.2020.9287105 fatcat:e7oqiqqlprellew2wcie7icz44

Fast and Accurate Optical Flow based Depth Map Estimation from Light Fields [article]

Yang Chen, Martin Alain, Aljosa Smolic
2020 arXiv   pre-print
Depth map estimation is a crucial task in computer vision, and new approaches have recently emerged taking advantage of light fields, as this new imaging modality captures much more information about the  ...  In this paper, we propose a novel depth estimation method from light fields based on existing optical flow estimation methods.  ...  We evaluate here the accuracy and efficiency of our proposed method against stateof-the-art light field depth map estimation methods [26, 11, 28, 25, 10, 22, 19] using the recent HCI 4D light field dataset  ... 
arXiv:2008.04673v1 fatcat:no6ekcoqwnfyfouo7ym3gyvq7i

Light Field Scale-Depth Space Transform for Dense Depth Estimation

Ivana Tosic, Kathrin Berkner
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops  
Experimental results on the HCI (Heidelberg Collaboratory for Image Processing) light field benchmark show that our method gives state of the art depth accuracy.  ...  We first propose a method for construction of light field scale-depth spaces, by convolving a given light field with a special kernel adapted to the light field structure.  ...  Conclusions We have presented a novel method for dense depth estimation from light fields, based on extrema detection in continuous light field scale-depth spaces built upon the normalized second derivative  ... 
doi:10.1109/cvprw.2014.71 dblp:conf/cvpr/TosicB14 fatcat:7om7hieds5ayhg2766hfru6bvi

𝕏Resolution Correspondence Networks [article]

Georgi Tinchev, Shuda Li, Kai Han, David Mitchell, Rigas Kouskouridas
2021 arXiv   pre-print
In this paper, we aim at establishing accurate dense correspondences between a pair of images with overlapping field of view under challenging illumination variation, viewpoint changes, and style differences  ...  Through an extensive ablation study of the state-of-the-art correspondence networks, we surprisingly discovered that the widely adopted 4D correlation tensor and its related learning and processing modules  ...  Acknowledgements We would like to thank Umar Ahmed, Guy Newsom and the rest of the XYZ Reality team for helping out with various design concepts and fruitful discussions.  ... 
arXiv:2012.09842v2 fatcat:urcvdjpzfbepnl4b5yd6u2bwou

Light Field Saliency Detection with Deep Convolutional Networks [article]

Jun Zhang, Yamei Liu, Shengping Zhang, Ronald Poppe, Meng Wang
2019 arXiv   pre-print
This makes our dataset suitable for training deeper networks and benchmarking. Furthermore, we propose a novel end-to-end CNN-based framework for light field saliency detection.  ...  Compared to current light field saliency datasets [1], [2], our new dataset is larger, of higher quality, contains more variation and more types of light field inputs.  ...  In order to directly synthesize novel views of dense 4D light fields from sparse views, Wang et al.  ... 
arXiv:1906.08331v2 fatcat:3xbuj443rvemdffwjdkw4garem

Compressive Light Field Reconstructions using Deep Learning [article]

Mayank Gupta, Arjun Jauhari, Kuldeep Kulkarni, Suren Jayasuriya, Alyosha Molnar, Pavan Turaga
2018 arXiv   pre-print
Light field imaging is limited in its computational processing demands of high sampling for both spatial and angular dimensions.  ...  We present a deep learning approach using a new, two branch network architecture, consisting jointly of an autoencoder and a 4D CNN, to recover a high resolution 4D light field from a single coded 2D image  ...  Acknowledgements: The authors would like to thank the anonymous reviewers for their detailed feedback, Siva Sankalp for running some experiments, and Mark Buckler for GPU computing support.  ... 
arXiv:1802.01722v1 fatcat:h4calra7hbd6vpqvs7yczbuhhm
« Previous Showing results 1 — 15 out of 1,840 results