Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
slfan da56239c9e | 7 months ago | |
---|---|---|
figures | 1 year ago | |
LICENSE | 1 year ago | |
README.md | 7 months ago |
A large-scale light field dataset developed for multiple research fields. If you find that our data violates your rights, please contact us immediately, and we will deal with it in time.
Compared with current RGB or RGB-D saliency detection datasets, those for light field saliency detection often suffer from many defects, e.g., insufficient data amount and diversity, incomplete data formats, and rough annotations, thus impeding the prosperity of this field. To settle these issues, we elaborately build a large-scale light field dataset, dubbed PKU-LF, comprising 5,000 light fields and covering diverse indoor and outdoor scenes. Our PKU-LF provides all-inclusive representation formats of light fields and offers a unified platform for comparing algorithms utilizing different input formats. For sparking new vitality in saliency detection tasks, we present many unexplored scenarios (such as underwater and high-resolution scenes) and the richest annotations (such as scribble annotations, bounding boxes, object-/instance-level annotations, and edge annotations), on which many potential attention modeling tasks can be investigated. To facilitate the development of saliency detection, we systematically evaluate and analyze 16 representative 2D, 3D, and 4D methods on four existing datasets and the proposed dataset, furnishing a thorough benchmark. Furthermore, tailored to the distinct structural characteristics of light fields, a novel symmetric two-stream architecture (STSA) network is proposed to predict the saliency of light fields more accurately. Specifically, our STSA incorporates a focalness interweavement module (FIM) and three partial decoder modules (PDM). The former is designed to efficiently establish long-range dependencies across focal slices, while the latter aims to effectively aggregate the extracted coadjutant features in a mutual-enhancement way. Extensive experiments demonstrate that our method can significantly outperform the competitors.
Figure 1: Illustration of light fields. (a) Theory of light fields. (b) Light field data in our PKU-LF.
We provide the results in our paper for a quick comparison, and you can download the predicted saliency maps of our method on different datasets from here.
Usage of our dataset is under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License.
Figure 2: Rich annotations of the proposed PKU-LF dataset including scribble annotations, bounding boxes, object-/instance-level annotations, and edge annotations.
Dataset | Calibration data | Raw data | All-focus image | Depth map | Focal stack | Relative depth of field | Sub-aperture image & Micro-lens image | Annotation |
---|---|---|---|---|---|---|---|---|
PKU-LF | Download | Download | Download | Download | Download | Download | Download | Download |
*Note: We made two major upgrades to the dataset. First, some annotation errors were corrected. Second, we expanded the number of samples to 9909 to meet the needs of existing learning paradigms for large-scale data.
Please cite our paper if you find our work is helpful.
@article{gao2023thorough,
title={A Thorough Benchmark and a New Model for Light Field Saliency Detection},
author={Gao, Wei and Fan, Songlin and Li, Ge and Lin, Weisi},
journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
year={2023},
publisher={IEEE}
}
A large-scale light field dataset developed for multiple research fields.
other
CC-BY-NC-ND-4.0
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》