OSCR

Volatile self-selective memristive neuron for millisecond-latency neuromorphic object detection at the edge.

Code ↔ Paper

The paper beside its authors' code: matches between them have not been computed for this paper yet.

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Markdown · 112 lines · 7.3 KB · no license

  1. # VisDrone-Dataset
  3. ![VisDrone](http://aiskyeye.com/upfile/1524040398110image_sample.jpg)
  5. Drones, or general UAVs, equipped with cameras have been fast deployed to a wide range of applications, including agricultural, aerial photography, fast delivery, and surveillance. Consequently, automatic understanding of visual data collected from these platforms become highly demanding, which brings computer vision to drones more and more closely. We are excited to present a large-scale benchmark with carefully annotated ground-truth for various important computer vision tasks, named VisDrone, to make vision meet drones. The VisDrone2019 dataset is collected by the AISKYEYE team at Lab of Machine Learning and Data Mining , Tianjin University, China. The benchmark dataset consists of 288 video clips formed by 261,908 frames and 10,209 static images, captured by various drone-mounted cameras, covering a wide range of aspects including location (taken from 14 different cities separated by thousands of kilometers in China), environment (urban and country), objects (pedestrian, vehicles, bicycles, etc.), and density (sparse and crowded scenes). Note that, the dataset was collected using various drone platforms (i.e., drones with different models), in different scenarios, and under various weather and lighting conditions. These frames are manually annotated with more than 2.6 million bounding boxes of targets of frequent interests, such as pedestrians, cars, bicycles, and tricycles. Some important attributes including scene visibility, object class and occlusion, are also provided for better data utilization.
  7. The challenge mainly focuses on four tasks:
  9. (1) Task 1: object detection in images challenge. The task aims to detect objects of predefined categories (e.g., cars and pedestrians) from individual images taken from drones.
  11. (2) Task 2: object detection in videos challenge. The task is similar to Task 1, except that objects are required to be detected from videos.
  13. (3) Task 3: single-object tracking challenge. The task aims to estimate the state of a target, indicated in the first frame, in the subsequent video frames.
  15. (4) Task 4: multi-object tracking challenge. The task aims to recover the trajectories of objects in each video frame.
  17. (5) Task 5: crowd counting challenge. The task aims to to count persons in each video frame.
  19. ## Download
  21. Note that the bounding box annotations of test-dev are avalialbe. Researchers can use test-dev to publish papers. testset-challenge is used for VisDrone2020 Challenge and the annotations is unavailable.
  23. ### Task 1: Object Detection in Images
  25. VisDrone-DET dataset
  27. * trainset (1.44 GB): [BaiduYun](https://pan.baidu.com/s/1K-JtLnlHw98UuBDrYJvw3A) | [GoogleDrive](https://drive.google.com/file/d/1a2oHjcEcwXP8oUF95qiwrqzACb2YlUhn/view?usp=sharing)
  29. * valset (0.07 GB): [BaiduYun](https://pan.baidu.com/s/1jdK_dAxRJeF2Xi50IoML1g) | [GoogleDrive](https://drive.google.com/file/d/1bxK5zgLn0_L8x276eKkuYA_FzwCIjb59/view?usp=sharing)
  31. * testset-dev (0.28 GB): [BaiduYun](https://pan.baidu.com/s/1RdRfSWV-1IFK7aWljLU_LQ) | [GoogleDrive](https://drive.google.com/open?id=1PFdW_VFSCfZ_sTSZAGjQdifF_Xd5mf0V) (GT avalialbe)
  33. * testset-challenge (0.28 GB): [BaiduYun](https://pan.baidu.com/s/1lvEkCgy1WWK4B7TLki4yBQ) | [GoogleDrive](https://drive.google.com/file/d/1KN8R3oioOvSXH492GEVk-Hx74nWHAcXT/view?usp=sharing)
  35. VisDrone-DET toolkit:
  37. * [Matlab beta](https://github.com/VisDrone/VisDrone2018-DET-toolkit)
  39. ### Task 2: Object Detection in Videos
  41. VisDrone-VID dataset
  43. * trainset (7.53 GB): [BaiduYun](https://pan.baidu.com/s/1kC3NTK6MPVv3D1CY9gXaCQ) | [GoogleDrive](https://drive.google.com/file/d/1NSNapZQHar22OYzQYuXCugA3QlMndzvw/view?usp=sharing)
  45. * valset (1.49 GB): [BaiduYun](https://pan.baidu.com/s/12-A6Mg1Gg7hyS4WwG27dDw) | [GoogleDrive](https://drive.google.com/file/d/1xuG7Z3IhVfGGKMe3Yj6RnrFHqo_d2a1B/view?usp=sharing)
  47. * testset-dev (2.14 GB): [BaiduYun](https://pan.baidu.com/s/1r1P5aJ1zOlQH_58LfYFzQQ) | [GoogleDrive](https://drive.google.com/open?id=1-BEq--FcjshTF1UwUabby_LHhYj41os5)(GT avalialbe)
  49. * testset-challenge (2.70 GB): [BaiduYun](https://pan.baidu.com/s/1ew6B-kvKV9yv__onnjA4dQ) | [GoogleDrive](https://drive.google.com/file/d/1Qwyp_cEpGyXGqJ8IbusEzuNHgbM403NP/view?usp=sharing)
  51. VisDrone-VID toolkit:
  53. * [Matlab beta](https://github.com/VisDrone/VisDrone2018-VID-toolkit)
  55. ### Task 3: Single-Object Tracking
  57. VisDrone-SOT dataset
  59. * trainset_part1 (7.78 GB): [BaiduYun](https://pan.baidu.com/s/1obWeT5DvBkTuBmO1NlSQ-Q) | [GoogleDrive](https://drive.google.com/file/d/1a-AmQjYATzj8seXLlEm9Sx8aQmClrJka/view?usp=sharing)
  61. * trainset_part2 (12.59 GB): [BaiduYun](https://pan.baidu.com/s/1c6iZeMJUXOIERxFJJ6B0jw) | [GoogleDrive](https://drive.google.com/file/d/16YPyhNDQrTgW8I2HaH_HNEO-KTRA-xso/view?usp=sharing)
  63. * valset (1.29 GB): [BaiduYun](https://pan.baidu.com/s/1WTWx4iyf33lnIyRu2uP_Hg) | [GoogleDrive](https://drive.google.com/file/d/18SNAOlCJtApnG2m45ud-1e_OtGYill0D/view?usp=sharing)
  65. * testset-dev (11.27 GB): [BaiduYun](https://pan.baidu.com/s/18j3umaWR_1fFy2ISOe9dGg) | [GoogleDrive](https://drive.google.com/open?id=1xCiHjU4JlR9QsYtiHYy2UUd3m6NthoBC)(GT avalialbe)
  67. * testset-challenge_part1 (17.40 GB): [BaiduYun](https://pan.baidu.com/s/14eWyaisDeciip-4_B14law) | [GoogleDrive](https://drive.google.com/file/d/1zxMLZrkkz4BkufbaDXQVMpDz-PRpPpxs/view?usp=sharing)
  69. * testset-challenge_part2 (17.31 GB): [BaiduYun](https://pan.baidu.com/s/1WDQ4JD5eLfRy-KVNyvKPig) | [GoogleDrive](https://drive.google.com/file/d/13J_9rwpf2TpVVsWgzMUv4uCxpOKDNbef/view?usp=sharing)
  71. * testset-challenge_initialization(12 KB): [BaiduYun](https://pan.baidu.com/s/1hvY0LXjTeuF8r-VVvVX1OQ) | [GoogleDrive](https://drive.google.com/file/d/1z7lQ0co8Dcu-ZojMN7dnIxn-i6QbGNL4/view?usp=sharing)
  73. VisDrone-SOT toolkit:
  75. * [Matlab beta](https://github.com/VisDrone/VisDrone2018-SOT-toolkit)
  77. ### Task 4: Multi-Object Tracking
  79. VisDrone-MOT dataset
  81. * trainset (7.53 GB): [BaiduYun](https://pan.baidu.com/s/16BtpKNWi0cEk8WUtfzpEHQ) | [GoogleDrive](https://drive.google.com/file/d/1-qX2d-P1Xr64ke6nTdlm33om1VxCUTSh/view?usp=sharing)
  83. * valset (1.48 GB): [BaiduYun](https://pan.baidu.com/s/1wTWFpHw4uLXPVCp1m5fQNQ) | [GoogleDrive](https://drive.google.com/file/d/1rqnKe9IgU_crMaxRoel9_nuUsMEBBVQu/view?usp=sharing)
  85. * testset-dev (2.14 GB): [BaiduYun](https://pan.baidu.com/s/1_gLvMxkMKb3RZjGyZv7btQ) | [GoogleDrive](https://drive.google.com/open?id=14z8Acxopj1d86-qhsF1NwS4Bv3KYa4Wu)(GT avalialbe)
  87. * testset-challenge (2.70 GB): [BaiduYun](https://pan.baidu.com/s/1xIloIRSj1FtcEoWI9esn7w) | [GoogleDrive](https://drive.google.com/file/d/1I0nn6dVKctzDE5YJ3q9qOlhKLiSIDAxF/view?usp=sharing)
  89. VisDrone-MOT toolkit:
  91. * [Matlab beta](https://github.com/VisDrone/VisDrone2018-MOT-toolkit)
  93. ### Task 5: Crowd Counting
  95. ECCV2020 Challenge
  96. DroneCrowd (1.03 GB): [BaiduYun](https://pan.baidu.com/share/init?surl=llJZJMi2L5oUQvj31iBlfg)(code: h0j8)|
  97. [GoogleDrive](https://drive.google.com/file/d/1HY3V4QObrVjzXUxL_J86oxn2bi7FMUgd/view?usp=sharing)
  100. ## Citation
  101. ```
  102. @article{zhu2021detection,
  103. title={Detection and tracking meet drones challenge},
  104. author={Zhu, Pengfei and Wen, Longyin and Du, Dawei and Bian, Xiao and Fan, Heng and Hu, Qinghua and Ling, Haibin},
  105. journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  106. volume={44},
  107. number={11},
  108. pages={7380--7399},
  109. year={2021},
  110. publisher={IEEE}
  111. }
  112. ```

README.md at commit 4364e82, no license · at the source

Overview

Authors: Zhejia Zhang1,2, Jiahua Xu3, Xuemeng Fan1,2, Guobin Zhang1,2,4, Zijian Wang1,2, Pengtao Li1,2,4, Qi Luo1,2, Haoxiang Yu1,2, Shuai Zhong5, Yunyan Zhang1,2, Wenzhang Fang1,2, Weidong You6, Daying Sun6, Kun Ren1,2, Qing Wan4, Yishu Zhang1,2,4
ORCID iDs: Zhejia Zhang, Jiahua Xu, Wenzhang Fang, Yishu Zhang
  1. College of Integrated Circuits, Zhejiang University, ZJU-Hangzhou Global Scientific and Technological Innovation Center, Hangzhou, China
  2. Zhejiang ICsprout Semiconductor Co., Ltd, Hangzhou, China
  3. The State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, China
  4. Yongjiang Laboratory, Ningbo, China
  5. Guangdong Institute of Intelligence Science and Technology, Hengqin, Zhuhai, China
  6. School of Microelectronics, Nanjing University of Science and Technology, Nanjing, China
Institutions: Zhejiang University (China); Xidian University (China); Yongjiang Laboratory (China); Guangdong Institute of Intelligence Science and Technology (China); Nanjing University of Science and Technology (China)
Journal: Nature communications, volume 17, issue 1, article 6951
Dates: received 12 December 2025; accepted 18 May 2026; published online 28 May 2026
Type: Research article · Language: English
License: CC BY-NC-ND
Identifiers: DOI 10.1038/s41467-026-73825-3 · PMID 42209522 · PMCID PMC13388706 · OpenAlex W7162642311
Open access: gold, a free copy (OpenAlex)
Status: empty repository
Categories: methods / tools (subfield)
Methods: Machine learning
Keywords: Electronic devices, Electrical and electronic engineering
Topic: Advanced Memory and Neural Computing (Electrical and Electronic Engineering, Engineering), according to OpenAlex
Funding: National Natural Science Foundation of China (National Science Foundation of China) (92364204, U25A20500)
Citations: not cited yet (Europe PMC); 57 references in the paper

Abstract

The abstract is not reproduced here: the paper's license (CC BY-NC-ND) does not allow it. Read it in the paper, at the publisher or on Europe PMC.

Repository

Its files are read in the Code ↔ Paper reader above.

VisDrone/VisDrone-Dataset

License: none: the authors keep all their rights
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: 4364e8265275dfa44fd8f767b5180af58580194d, 24 September 2023
Size: 1 file, 0 scripts
Software Heritage: not archived
Found in: “Code availability”
Holds: README
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
1 file
At the source: github.com/VisDrone/VisDrone-Dataset

Code availability statement

The paper has a code availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

Read it in the paper: doi.org/10.1038/s41467-026-73825-3.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 0 scripts, each with its path and the digest of its content;
  • no match between paragraphs and code yet;
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data availability statement

The paper has a data availability statement. Its license (CC BY-NC-ND) does not allow reproducing it here; in short, from what the harvester recognized in it:

  • no repository, dataset or request procedure was recognized in it

Read it in the paper: doi.org/10.1038/s41467-026-73825-3.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 16 authors, 2 keywords, 1 funder, 29 references.

Cite

This paper

Zhang, Z., Xu, J., Fan, X., Zhang, G., Wang, Z., Li, P., Luo, Q., Yu, H., Zhong, S., Zhang, Y., Fang, W., You, W., Sun, D., Ren, K., Wan, Q., & Zhang, Y. (2026). Volatile self-selective memristive neuron for millisecond-latency neuromorphic object detection at the edge. Nature communications, 17(1), 6951. https://doi.org/10.1038/s41467-026-73825-3

BibTeX

@article{zhang2026volatile,
author = {Zhang, Zhejia and Xu, Jiahua and Fan, Xuemeng and Zhang, Guobin and Wang, Zijian and Li, Pengtao and Luo, Qi and Yu, Haoxiang and Zhong, Shuai and Zhang, Yunyan and Fang, Wenzhang and You, Weidong and Sun, Daying and Ren, Kun and Wan, Qing and Zhang, Yishu},
title = {{Volatile self-selective memristive neuron for millisecond-latency neuromorphic object detection at the edge}},
journal = {Nature communications},
year = {2026},
month = may,
volume = {17},
number = {1},
pages = {6951},
publisher = {Nature Publishing Group},
issn = {2041-1723},
doi = {10.1038/s41467-026-73825-3},
url = {https://doi.org/10.1038/s41467-026-73825-3},
pmid = {42209522},
pmcid = {PMC13388706}
}

RIS

TY - JOUR
AU - Zhang, Zhejia
AU - Xu, Jiahua
AU - Fan, Xuemeng
AU - Zhang, Guobin
AU - Wang, Zijian
AU - Li, Pengtao
AU - Luo, Qi
AU - Yu, Haoxiang
AU - Zhong, Shuai
AU - Zhang, Yunyan
AU - Fang, Wenzhang
AU - You, Weidong
AU - Sun, Daying
AU - Ren, Kun
AU - Wan, Qing
AU - Zhang, Yishu
TI - Volatile self-selective memristive neuron for millisecond-latency neuromorphic object detection at the edge
T2 - Nature communications
J2 - Nat Commun
PY - 2026
DA - 2026/05/28
VL - 17
IS - 1
SP - 6951
SN - 2041-1723
PB - Nature Publishing Group
DO - 10.1038/s41467-026-73825-3
UR - https://doi.org/10.1038/s41467-026-73825-3
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41467-026-73825-3",
"type": "article-journal",
"title": "Volatile self-selective memristive neuron for millisecond-latency neuromorphic object detection at the edge",
"container-title": "Nature communications",
"author": [
{
"family": "Zhang",
"given": "Zhejia"
},
{
"family": "Xu",
"given": "Jiahua"
},
{
"family": "Fan",
"given": "Xuemeng"
},
{
"family": "Zhang",
"given": "Guobin"
},
{
"family": "Wang",
"given": "Zijian"
},
{
"family": "Li",
"given": "Pengtao"
},
{
"family": "Luo",
"given": "Qi"
},
{
"family": "Yu",
"given": "Haoxiang"
},
{
"family": "Zhong",
"given": "Shuai"
},
{
"family": "Zhang",
"given": "Yunyan"
},
{
"family": "Fang",
"given": "Wenzhang"
},
{
"family": "You",
"given": "Weidong"
},
{
"family": "Sun",
"given": "Daying"
},
{
"family": "Ren",
"given": "Kun"
},
{
"family": "Wan",
"given": "Qing"
},
{
"family": "Zhang",
"given": "Yishu"
}
],
"container-title-short": "Nat Commun",
"volume": "17",
"issue": "1",
"page": "6951",
"DOI": "10.1038/s41467-026-73825-3",
"PMID": "42209522",
"PMCID": "PMC13388706",
"ISSN": "2041-1723",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41467-026-73825-3",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
28
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1038/s41467-026-72428-2 [code]
Artificial plateau neurons with in-situ spike-malleability for rhythmic quadrupedal locomotion.
Journal: Nature communications
In common: 2 references, author Yishu Zhang
[2] doi:10.1021/acs.chemrev.5c00878
Self-Oscillatory Neuron-like Devices for Unconventional Computing Applications.
Journal: Chemical reviews
In common: 3 references
[3] doi:10.1038/s41467-026-72900-z [code]
Resistive memory-based neural differential equation solver for score-based diffusion model.
Journal: Nature communications
In common: 3 references
[4] doi:10.3390/s26061801 [code]
Characterization of a Spiking Convolutional Processor for FPGA.
Journal: Sensors (Basel, Switzerland)
In common: methods / tools, 2 references
[5] doi:10.1038/s41467-026-75488-6
Realization of the Bienenstock-Cooper-Munro rule in a single memristor.
Journal: Nature communications
In common: 2 references
[6] doi:10.1021/acs.jpclett.6c01632
Spiking without Resets: Continuous Integrate-and-Fire Dynamics in Neuronal Circuits.
Journal: The journal of physical chemistry letters
In common: 2 references
[7] doi:10.1371/journal.pcbi.1014432 [code]
Spiking neurons as predictive controllers of linear systems.
Journal: PLoS computational biology
In common: 1 reference
[8] doi:10.3390/s26134232
HRL-Det: Hierarchical Reinforcement Learning for Sequential Object Detection in Aerial Imagery.
Journal: Sensors (Basel, Switzerland)
In common: 1 reference
[9] doi:10.1038/s41467-026-74769-4 [code]
Amphibian-inspired neuromorphic dynamic vision systems based on ferroelectric field-effect transistor.
Journal: Nature communications
In common: 1 reference
[10] doi:10.3389/fnbot.2026.1761767 [code]
Trainable movement control using spikes and muscle-twitch dynamics.
Journal: Frontiers in neurorobotics
In common: 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.