UP-COUNT Dataset

High-resolution dataset for people localisation from UAV with moving camera, altitude information and counting diversity. Test videos along with tracking labels you can find in UP-COUNT-TRACK dataset.

Paper (preprint)

Data

Code

PUT Vision Lab


This dataset is a part of the "Enhancing people localisation in drone imagery for better crowd management by utilising every pixel in high-resolution images" paper.

202 Sequences
10'000 Frames
352'487 Labels
3840 x 2160 px

Abstract

Accurate people localisation using drones is crucial for effective crowd management, not only during massive events and public gatherings but also for monitoring daily urban crowd flow. Traditional methods for tiny object localisation using high-resolution drone imagery often face limitations in precision and efficiency, primarily due to constraints in image scaling and sliding window techniques. To address these challenges, a novel approach dedicated to point-oriented object localisation is proposed. Along with this approach, the Pixel Distill module is introduced to enhance the processing of high-definition images by extracting spatial information from individual pixels at once. Additionally, a new dataset named Up-Count, tailored to contemporary drone applications, is shared. It addresses a wide range of challenges in drone imagery, such as simultaneous camera and object movement during the image acquisition process, pushing forward the capabilities of crowd management applications. A comprehensive evaluation of the proposed method on the proposed dataset and the commonly used DroneCrowd dataset demonstrates the superiority of our approach over existing methods and highlights its efficacy in drone-based crowd object localisation tasks. These improvements markedly increase the algorithm's applicability to operate in real-world scenarios, enabling more reliable localisation and counting of individuals in dynamic environments.

Features

Label counting diversity
Significant label count variability: 0 - 1039 (mean: 35.25)


Moving camera
Camera movement during the image acquisition process


Drone altitude
Lowest altitude: 26.0 meters | Highest altitude: 101.0 meters


High resolution
High frames resolution: 3840 x 2160 px



Download dataset

Note: The dataset is licensed under the Creative Commons Attribution Non-Commercial 4.0 International.
If you want to use the dataset commercially, don't hesitate to contact the authors: vision@put.poznan.pl.



Samples

Examples of each sequence (GDrive) | Example of full sequence (GDrive)



Full dataset

GDrive | Zenodo | PUT Cloud



Citation

If you find our work valuable, please cite:
@article{ptak2025enhancing,
title={Enhancing people localisation in drone imagery for better crowd management by utilising every pixel in high-resolution images},
author={Ptak, Bartosz and Kraft, Marek},
journal={arXiv preprint arXiv:2502.04014},
year={2025} }

Most crowed sequence



The lowest alitude



The highest altitude



Example dataset usage in an application





Sequences statistics

Sequence ID Subset Frames length Average altitude Average counting Counting sum Counting min Counting max
0000 train 46 60.0 21.9 1007 14 35
0001 test 90 61.4 104.3 9387 66 160
0002 val 28 74.7 73.9 2068 47 93
0003 train 20 62.0 72.2 1443 61 81
0004 train 22 84.1 97.0 2133 68 124
0005 val 46 90.1 50.3 2313 25 86
0006 train 33 63.5 65.2 2150 18 112
0007 test 115 77.7 6.9 797 0 13
0008 train 52 65.3 15.2 788 3 23
0009 train 32 76.2 12.0 384 10 15
0010 train 135 76.0 18.2 2453 6 45
0011 train 27 67.3 21.0 567 8 51
0012 train 67 86.3 32.3 2162 8 93
0013 train 31 68.5 17.5 542 8 32
0014 test 88 68.2 37.1 3263 8 68
0015 train 95 80.2 715.3 67955 522 864
0016 test 24 50.7 970.3 23287 850 1039
0017 train 51 30.2 7.7 394 0 13
0018 train 49 27.0 6.3 311 0 14
0019 train 31 31.0 6.9 214 0 11
0020 train 21 30.8 3.7 77 0 8
0021 train 22 30.7 5.0 109 2 8
0022 train 31 30.7 1.6 51 0 6
0023 train 16 30.7 9.9 158 1 20
0024 val 32 30.8 15.4 492 1 23
0025 val 17 30.8 23.1 392 4 34
0026 train 38 30.9 14.8 562 4 24
0027 train 29 31.0 29.1 844 3 55
0028 test 23 31.0 24.7 569 5 45
0029 train 20 30.9 10.7 213 0 22
0030 train 34 30.0 1.7 57 0 3
0031 train 35 30.0 8.8 308 0 20
0032 val 16 30.0 13.2 212 3 19
0033 train 27 28.7 5.8 157 0 14
0034 train 29 30.0 7.9 228 3 15
0035 test 24 30.0 9.2 220 0 12
0036 train 19 30.0 7.8 149 0 14
0037 test 53 30.2 9.5 506 0 31
0038 val 39 30.3 10.6 414 0 26
0039 test 45 30.3 11.3 508 0 31
0040 train 39 30.2 23.4 913 0 49
0041 train 29 30.4 11.8 341 6 17
0042 train 54 28.3 18.1 980 2 41
0043 val 33 28.3 24.3 802 10 41
0044 train 14 47.1 22.3 312 6 36
0045 train 34 45.4 17.1 580 9 29
0046 test 48 49.7 10.5 504 1 13
0047 train 42 49.7 14.7 619 6 19
0048 test 35 49.6 8.9 312 2 17
0049 val 7 49.6 6.6 46 4 8
0050 test 52 49.6 6.8 351 5 11
0051 train 26 49.5 9.9 257 5 14
0052 val 32 49.6 6.6 211 3 10
0053 train 44 50.5 21.4 941 10 33
0054 train 34 50.6 17.5 595 8 28
0055 train 50 51.0 12.2 608 6 21
0056 train 11 51.1 21.2 233 15 26
0057 train 46 49.5 13.2 609 7 26
0058 val 32 49.3 8.3 267 0 15
0059 test 42 49.5 15.5 649 5 24
0060 train 36 49.5 23.4 843 7 31
0061 train 3 48.5 30.0 90 27 32
0062 test 24 48.0 17.5 421 2 31
0063 train 33 30.1 1.3 43 0 5
0064 train 62 30.0 5.8 358 1 13
0065 train 32 29.7 9.5 303 2 17
0066 test 50 28.9 12.0 600 0 20
0067 train 18 29.8 16.9 305 0 30
0068 train 65 30.2 4.9 318 1 9
0069 val 62 30.1 3.2 196 0 9
0070 train 14 30.0 21.9 307 3 30
0071 train 23 29.9 16.4 378 2 33
0072 train 78 49.5 22.5 1753 6 37
0073 train 50 48.8 16.8 839 6 24
0074 train 27 48.8 9.3 251 8 11
0075 val 37 48.8 33.5 1240 21 45
0076 train 25 48.7 32.2 804 14 43
0077 train 18 48.7 36.3 653 17 50
0078 val 90 30.1 10.4 937 1 30
0079 train 49 30.1 10.4 510 3 23
0080 train 21 30.2 11.0 231 3 18
0081 test 46 30.6 15.6 717 2 37
0082 train 5 30.6 3.0 15 2 4
0083 train 42 30.6 20.9 878 2 46
0084 test 43 30.6 33.4 1438 0 73
0085 train 19 30.6 16.6 315 2 37
0086 test 19 30.6 39.7 755 12 56
0087 train 53 28.7 9.7 515 5 21
0088 val 59 28.9 9.4 552 2 21
0089 train 61 30.5 10.2 623 3 20
0090 val 108 30.4 7.0 751 0 14
0091 train 88 30.5 9.3 820 3 17
0092 train 89 30.9 8.7 775 1 18
0093 train 105 31.0 7.0 735 1 16
0094 val 78 30.9 16.0 1245 11 21
0095 train 53 29.9 22.5 1191 6 30
0096 train 60 29.9 13.6 817 3 23
0097 train 29 29.4 19.7 572 9 27
0098 train 99 60.2 15.9 1572 6 26
0099 test 68 79.6 23.7 1610 13 37
0100 train 70 70.0 30.3 2121 19 45
0101 val 75 67.3 27.0 2026 18 37
0102 train 74 98.7 57.9 4283 31 81
0103 train 89 99.4 60.7 5404 14 113
0104 test 84 99.4 47.3 3977 24 66
0105 train 64 99.4 38.2 2445 22 49
0106 test 33 100.4 13.2 437 4 23
0107 train 45 100.4 32.1 1445 18 44
0108 val 53 100.4 68.5 3632 5 139
0109 train 43 79.8 73.0 3139 25 109
0110 train 58 79.7 51.3 2974 13 88
0111 val 45 80.0 51.4 2311 16 85
0112 test 59 80.1 62.8 3704 13 124
0113 train 35 64.6 17.0 596 1 25
0114 train 50 97.0 34.0 1702 12 51
0115 train 89 80.1 40.3 3586 2 85
0116 test 56 100.7 62.4 3496 14 128
0117 train 41 50.3 27.7 1135 9 53
0118 train 27 50.2 25.7 693 0 61
0119 val 41 50.2 18.5 759 0 34
0120 train 22 100.2 62.4 1372 49 76
0121 train 37 100.3 68.6 2539 34 86
0122 train 68 100.2 76.9 5227 9 112
0123 val 25 100.0 88.6 2214 64 100
0124 train 74 60.5 52.3 3869 7 97
0125 test 39 60.6 31.2 1215 20 43
0126 train 36 59.9 23.9 861 11 32
0127 val 43 69.8 19.3 831 9 29
0128 train 63 69.2 29.9 1883 9 39
0129 train 36 79.2 43.0 1547 32 52
0130 train 70 80.2 37.0 2591 9 64
0131 test 58 100.7 8.7 504 1 19
0132 val 63 100.7 7.1 445 2 12
0133 train 2 71.6 16.0 32 16 16
0134 train 33 100.3 19.0 628 5 24
0135 train 37 100.3 15.6 577 3 22
0136 train 26 100.3 15.4 400 2 26
0137 train 50 100.6 17.9 895 1 35
0138 train 67 70.2 18.9 1263 8 32
0139 train 34 100.6 39.2 1333 24 49
0140 train 38 100.6 32.7 1244 18 46
0141 train 35 100.4 39.1 1368 17 66
0142 train 30 100.4 39.8 1195 24 52
0143 val 35 100.5 30.7 1074 23 38
0144 train 49 100.4 37.5 1839 26 46
0145 train 24 60.0 18.7 448 6 28
0146 test 43 59.7 10.3 443 1 26
0147 train 23 59.8 13.8 317 6 30
0148 train 29 59.8 17.1 495 7 23
0149 val 31 60.0 12.2 379 9 15
0150 train 49 69.6 11.7 575 3 19
0151 train 33 69.9 13.4 443 7 20
0152 train 55 70.0 26.5 1456 2 59
0153 val 63 80.3 35.7 2247 21 55
0154 test 44 80.3 19.1 840 5 27
0155 train 35 60.5 13.6 477 3 32
0156 train 23 60.7 5.0 115 2 8
0157 train 57 80.4 6.9 395 2 26
0158 train 144 80.0 7.2 1033 0 24
0159 test 70 61.0 11.3 789 4 18
0160 train 33 60.9 8.9 293 3 17
0161 train 112 30.5 5.6 629 0 14
0162 test 87 30.5 8.2 716 2 20
0163 train 53 30.5 12.4 655 0 39
0164 train 74 30.5 13.2 975 3 28
0165 val 68 100.1 22.0 1494 9 38
0166 train 68 100.1 22.7 1542 12 46
0167 train 63 79.8 24.0 1514 10 35
0168 train 90 79.8 18.7 1685 2 40
0169 train 60 79.8 9.4 564 4 14
0170 train 76 100.1 13.8 1047 5 22
0171 train 64 100.0 11.2 714 3 18
0172 train 101 100.0 14.5 1464 6 19
0173 train 35 60.7 4.9 172 1 10
0174 train 52 60.6 5.0 258 1 10
0175 train 24 61.0 18.3 440 12 22
0176 train 42 61.0 22.0 922 5 36
0177 train 33 80.8 40.3 1331 20 48
0178 train 32 80.9 29.4 942 13 41
0179 test 76 88.6 16.1 1224 7 25
0180 val 66 59.9 10.2 676 2 19
0181 val 29 53.1 130.2 3777 83 222
0182 train 45 49.7 125.4 5643 72 200
0183 train 60 52.3 132.4 7945 65 184
0184 test 128 70.4 36.4 4659 8 75
0185 train 106 52.5 13.5 1436 0 38
0186 train 19 30.7 4.1 78 2 7
0187 train 43 31.1 11.4 489 5 22
0188 train 67 30.9 11.6 777 2 24
0189 train 77 99.8 112.8 8689 27 193
0190 train 33 99.8 144.3 4762 97 299
0191 train 115 46.3 32.8 3770 8 59
0192 train 71 46.4 31.1 2205 8 82
0193 train 102 47.0 35.5 3618 15 61
0194 train 112 46.9 33.1 3702 11 57
0195 train 40 30.4 15.2 609 3 29
0196 test 50 30.4 15.7 787 1 30
0197 train 178 62.6 42.3 7533 5 75
0198 train 71 62.6 25.0 1774 11 42
0199 train 39 79.6 22.5 876 16 30
0200 train 47 62.5 107.2 5040 41 165
0201 val 38 62.2 14.4 548 4 24