154:
162:
146:
130:
125:) that help maintain a particular stable position during their flight. In windy conditions, the drone automatically makes fine moves to maintain its position and that changes the view near the boundary. It may be possible that some new objects appear near the image boundary. Overall, these affect classification, detection, and eventually tracking accuracy.
234:
State-of-the-art object detectors allow only the fixed size of image and change the input image size according to it. This change may deform the small objects in the image. The tiling approach helps when an image has a high resolution than the model's fixed input size; instead of scaling it down, the
186:
model's output depends on "How well it is trained." So, the data set must include small objects to detect such objects. Also, modern-day detectors, such as YOLO, rely on anchors. Latest versions of YOLO (starting from YOLOv5) uses an auto-anchor algorithm to find good anchors based on the nature of
727:
Li, Chuyi; Li, Lulu; Jiang, Hongliang; Weng, Kaiheng; Geng, Yifei; Li, Liang; Ke, Zaidan; Li, Qingyuan; Cheng, Meng; Nie, Weiqiang; Li, Yiduo; Zhang, Bo; Liang, Yufei; Zhou, Linyuan; Xu, Xiaoming (2022-09-07). "YOLOv6: A Single-Stage Object
Detection Framework for Industrial Applications".
256:
Instead of modifying existing methods, some add-on techniques are there, which can be directly placed on top of existing approaches to detect smaller objects. One such technique is
Slicing Aided Hyper Inference(SAHI). The image is sliced into different-sized multiple overlapping patches.
43:
269:
Various deep learning techniques are available that focus on such object detection problems: e.g., Feature-Fused SSD, YOLO-Z. Such methods work on "How to sustain features of small objects while they pass through convolution networks."
247:
network to learn features at a multi-scale: e.g., Twin
Feature Pyramid Networks (TFPN), Extended Feature Pyramid Network (EFPN). FPN helps to sustain features of small objects against convolution layers.
105:
Sometimes, the shadow of an object is detected as a part of object itself. So, the placement of the bounding box tends to centre around a shadow rather than an object. In the case of vehicle detection,
1240:
26:
where various techniques are employed to detect small objects in digital images and videos. "Small objects" are objects having a small pixel footprint in the input image. In areas such as
48:
47:
44:
49:
198:
models have billions of neurons that settle down to some weights after training. Therefore, it requires a good amount of quantitative and qualitative data for better training.
1853:
1883:
1805:
749:
Wang, Chien-Yao; Bochkovskiy, Alexey; Liao, Hong-Yuan Mark (2022-07-06). "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors".
222:
Selecting anchor size plays a vital role in small object detection. Instead of hand picking it, use algorithms that identify it based on the data set. YOLOv5 uses a
997:
2177:
1878:
261:
define their dimensions. Then patches are resized, while maintaining the aspect ratio during fine-tuning. These patches are then provided for training the model.
46:
153:
1302:
Benjumea, Aduen; Teeti, Izzeddin; Cuzzolin, Fabio; Bradley, Andrew (2021-12-23). "YOLO-Z: Improving small object detection in YOLOv5 for autonomous vehicles".
1101:
Lin, Tsung-Yi; Dollár, Piotr; Girshick, Ross; He, Kaiming; Hariharan, Bharath; Belongie, Serge (2017-04-19). "Feature
Pyramid Networks for Object Detection".
102:. As an object passes through convolution layers, its size gets reduced. Therefore, the small object disappears after several layers and becomes undetectable.
2110:
1843:
1838:
1188:
Akyon, Fatih
Cagatay; Altinuc, Sinan Onur; Temizel, Alptekin (2022-07-12). "Slicing Aided Hyper Inference and Fine-Tuning for Small Object Detection".
1873:
1941:
1958:
954:
Jocher, Glenn; Chaurasia, Ayush; Stoken, Alex; Borovec, Jirka; NanoCode012; Kwon, Yonghye; TaoXie; Michael, Kalen; Fang, Jiacong (2022-08-17).
1978:
1663:
1618:
1573:
1479:
1434:
1389:
1344:
1215:
1143:
1077:
835:
791:
439:
353:
1603:
2020 17th
International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON)
1888:
1858:
1848:
1828:
1798:
622:
Redmon, Joseph; Divvala, Santosh; Girshick, Ross; Farhadi, Ali (2016-05-09). "You Only Look Once: Unified, Real-Time Object
Detection".
161:
145:
2043:
1868:
1898:
1278:
313:
258:
210:
These help to get more features from objects and eventually learn the best from them. For example, a bike object in the 1280 X 1280
45:
1167:
Deng, Chunfang; Wang, Mengmeng; Liu, Liang; Liu, Yong (2020-04-09). "Extended
Feature Pyramid Network for Small Object Detection".
685:
Bochkovskiy, Alexey; Wang, Chien-Yao; Liao, Hong-Yuan Mark (2020-04-22). "YOLOv4: Optimal Speed and
Accuracy of Object Detection".
1552:
Santhanam, Sanjay; B, Sudhir
Sidhaarthan; Panigrahi, Sai Sudha; Kashyap, Suryakant Kumar; Duriseti, Bhargav Krishna (2021-11-26).
2048:
2033:
1791:
1011:
706:
Wang, Chien-Yao; Bochkovskiy, Alexey; Liao, Hong-Yuan Mark (2021-02-21). "Scaled-YOLOv4: Scaling Cross Stage Partial Network".
303:
2142:
2073:
1924:
62:
298:
1893:
956:"ultralytics/yolov5: v6.2 - YOLOv5 Classification Models, Apple M1, Reproducibility, ClearML and Deci.ai integrations"
1035:
Zhong, Yuanyi; Wang, Jianfeng; Peng, Jian; Zhang, Lei (2020-01-26). "Anchor Box Optimization for Object Detection".
2038:
308:
244:
2090:
2003:
99:
2194:
2068:
235:
image is broken down into tiles and then used in training. The same approach is used during inference as well.
1464:
2021 IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER)
2130:
2120:
1863:
114:
174:
There are various ways to detect small objects with existing techniques. Some of them are mentioned below,
2214:
2167:
1914:
2172:
1983:
1929:
977:
70:
776:
2020 IEEE 9th Joint International Information Technology and Artificial Intelligence Conference (ITAIC)
385:. Proceedings of International Conference on Communication, Computing and Virtualization (ICCCV) 2016.
574:
378:
2095:
2078:
2058:
2028:
1258:
519:
1370:"Deep Transfer Learning Enabled High-Density Crowd Detection and Classification using Aerial Images"
2100:
1963:
187:
object sizes in the data set. Therefore, it is mandatory to have smaller objects in the data set.
2115:
2023:
2008:
1968:
1724:
1669:
1624:
1579:
1534:
1485:
1440:
1395:
1350:
1303:
1284:
1248:
1221:
1193:
1168:
1149:
1122:
Liang, Yi; Changjian, Wang; Fangzhao, Li; Yuxing, Peng; Qin, Lv; Yuan, Yuan; Zhen, Huang (2019).
1102:
1083:
1036:
936:
908:
841:
797:
750:
729:
707:
686:
665:
644:
623:
604:
488:
445:
359:
223:
95:
27:
1780:
dataset by AISKYEYE team at Lab of Machine Learning and Data Mining, Tianjin University, China.
1558:
2021 International Conference on Computational Intelligence and Computing Applications (ICCICA)
2053:
1998:
1950:
1716:
1708:
1659:
1614:
1569:
1526:
1475:
1430:
1385:
1340:
1274:
1239:
Cao, Guimei; Xie, Xuemei; Yang, Wenzhe; Liao, Quan; Shi, Guangming; Wu, Jinjian (2018-04-10).
1211:
1139:
1073:
928:
831:
787:
596:
555:
537:
480:
435:
400:
349:
199:
66:
1325:"Bigdata Enabled Realtime Crowd Surveillance Using Artificial Intelligence and Deep Learning"
2219:
2162:
2125:
1973:
1833:
1755:
1700:
1651:
1606:
1561:
1516:
1467:
1422:
1377:
1332:
1266:
1203:
1131:
1065:
959:
918:
872:
823:
779:
586:
545:
527:
472:
427:
390:
341:
211:
183:
31:
23:
2157:
2105:
1814:
990:
142:
Various methods are available to detect small objects, which fall under three categories:
1419:
2018 International Conference on Current Trends towards Converging Technologies (ICCTCT)
1262:
523:
420:"Learning Multi-view Deep Features for Small Object Retrieval in Surveillance Scenarios"
2182:
2152:
2063:
1988:
1919:
1610:
1471:
550:
507:
82:
1374:
2022 6th International Conference on Computing Methodologies and Communication (ICCMC)
2208:
2147:
1628:
1583:
1565:
1538:
1489:
1399:
1354:
1225:
1153:
1128:
2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI)
1087:
1062:
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
940:
801:
608:
492:
195:
74:
1673:
1444:
1381:
1336:
1288:
845:
783:
426:. MM '15. New York, NY, USA: Association for Computing Machinery. pp. 859–862.
363:
2085:
1728:
1207:
449:
338:
2015 International Conference on Control Communication & Computing India (ICCC)
78:
58:
827:
664:
Redmon, Joseph; Farhadi, Ali (2018-04-08). "YOLOv3: An Incremental Improvement".
643:
Redmon, Joseph; Farhadi, Ali (2016-12-25). "YOLO9000: Better, Faster, Stronger".
1993:
1655:
1521:
1426:
923:
395:
1777:
1688:
1643:
1598:
1553:
1504:
1459:
1414:
1369:
1324:
1123:
1057:
896:
815:
771:
591:
476:
418:
Guo, Haiyun; Wang, Jinqiao; Xu, Min; Zha, Zheng-Jun; Lu, Hanqing (2015-10-13).
333:
1644:"Automatic detection of moving wild animals in airborne remote sensing images"
345:
107:
1712:
1530:
1460:"Crowd Counting in High Dense Images using Deep Convolutional Neural Network"
1135:
1069:
932:
600:
541:
484:
464:
404:
202:
is useful technique to generate more diverse data from an existing data set.
1458:
Sharath, S.V.; Biradar, Vidyadevi; Prajwal, M.S.; Ashwini, B. (2021-11-19).
1329:
2021 IEEE International Conference on Big Data and Smart Computing (BigComp)
508:"Real-Time Abnormal Object Detection for Video Surveillance in Smart Cities"
431:
122:
1760:
1743:
1720:
1704:
1245:
Ninth International Conference on Graphic and Image Processing (ICGIP 2017)
963:
877:
860:
559:
419:
117:
are very widely used in aerial imagery. They are equipped with hardware (
129:
57:
Small object detection has applications in various fields such as Video
1270:
532:
34:
object detection techniques under performed because of small objects.
1599:"Detection of Animal Behind Cages Using Convolutional Neural Network"
1368:
Sivachandiran, S.; Mohan, K. Jagan; Nazer, G. Mohammed (2022-03-29).
816:"Interactive workshop "How drones are changing the world we live in""
465:"Person retrieval in surveillance videos using attribute recognition"
118:
955:
1308:
1253:
1198:
1173:
1107:
1041:
913:
861:"An Evaluation of Deep Learning Methods for Small Object Detection"
770:
Zhang, Mingrui; Zhao, Wenbing; Li, Xiying; Wang, Dan (2020-12-11).
755:
734:
712:
691:
670:
649:
628:
1783:
424:
Proceedings of the 23rd ACM international conference on Multimedia
334:"Traffic video surveillance: Vehicle detection and classification"
160:
152:
144:
128:
41:
859:
Nguyen, Nhat-Duy; Do, Tien; Ngo, Thanh Duc; Le, Duy-Dinh (2020).
820:
2016 Integrated Communications Navigation and Surveillance (ICNS)
772:"Shadow Detection of Moving Objects in Traffic Monitoring Video"
206:
Increasing image capture resolution and model’s input resolution
1787:
1648:
2010 IEEE International Geoscience and Remote Sensing Symposium
1693:
IEEE Transactions on Pattern Analysis and Machine Intelligence
1505:"A Survey of Vehicle Re-Identification Based on Deep Learning"
463:
Galiyawala, Hiren; Raval, Mehul S.; Patel, Meet (2022-05-20).
1190:
2022 IEEE International Conference on Image Processing (ICIP)
1056:
Unel, F. Ozge; Ozkalayci, Burak O.; Cigla, Cevahir (2019).
778:. Vol. 9. Chongqing, China: IEEE. pp. 1983–1987.
1597:
Li, Nopparut; Kusakunniran, Worapan; Hotta, Seiji (2020).
1124:"TFPN: Twin Feature Pyramid Networks for Object Detection"
1742:
Cui, Suxia; Zhou, Yu; Wang, Yonghui; Zhai, Lujun (2020).
1012:"The Size and Quality of a Data Set | Machine Learning"
469:
Journal of Ambient Intelligence and Humanized Computing
214:
image has more features than the 640 X 640 resolution.
1554:"Animal Detection for Road safety using Deep Learning"
1331:. Jeju Island, Korea (South): IEEE. pp. 129–132.
1748:
Applied Computational Intelligence and Soft Computing
1241:"Feature-fused SSD: Fast detection for small objects"
379:"Automatic Traffic Surveillance Using Video Tracking"
340:. Trivandrum, Kerala, India: IEEE. pp. 516–521.
265:
Well-Optimised techniques for small object detection
1949:
1940:
1907:
1821:
506:Ingle, Palash Yuvraj; Kim, Young-Gab (2022-05-19).
1415:"Crowd Scene Analysis Using Deep Learning Network"
573:Tsuboi, Tsutomu; Yoshikawa, Noriaki (2020-03-01).
191:Generating more data via augmentation, if required
895:Gong, Zhiqiang; Zhong, Ping; Hu, Weidong (2019).
110:and two-wheeler detection suffer because of this.
1687:Ramanan, D.; Forsyth, D.A.; Barnard, K. (2006).
1058:"The Power of Tiling for Small Object Detection"
1130:. Portland, OR, USA: IEEE. pp. 1702–1707.
1064:. Long Beach, CA, USA: IEEE. pp. 582–591.
94:Modern-day object detection algorithms such as
1323:Rajendran, Logesh; Shyam Shankaran, R (2021).
865:Journal of Electrical and Computer Engineering
1799:
1503:Wang, Hongbo; Hou, Jiaying; Chen, Na (2019).
230:Tiling approach during training and inference
8:
1605:. Phuket, Thailand: IEEE. pp. 242–245.
575:"Traffic flow analysis in Ahmedabad (India)"
1247:. Vol. 10615. SPIE. pp. 381–388.
1946:
1806:
1792:
1784:
1376:. Erode, India: IEEE. pp. 1313–1317.
822:. Herndon, VA: IEEE. 2016. pp. 1–17.
178:Choosing a data set that has small objects
1759:
1520:
1307:
1252:
1197:
1172:
1106:
1040:
922:
912:
876:
754:
733:
711:
690:
669:
648:
627:
590:
549:
531:
394:
98:heavily uses convolution layers to learn
16:Detecting small objects in digital images
996:CS1 maint: numeric names: authors list (
1689:"Building models of animals from video"
324:
1959:3D reconstruction from multiple images
1466:. Nitte, India: IEEE. pp. 30–34.
986:
975:
1979:Simultaneous localization and mapping
1642:Oishi, Yu; Matsunaga, Tsuneo (2010).
1560:. Nagpur, India: IEEE. pp. 1–5.
890:
888:
7:
1744:"Fish Detection Using Deep Learning"
2044:Automatic number-plate recognition
1611:10.1109/ECTI-CON49241.2020.9158137
1472:10.1109/DISCOVER52564.2021.9663716
1413:Santhini, C.; Gomathi, V. (2018).
1243:. In Dong, Junyu; Yu, Hui (eds.).
14:
314:Intelligent transportation system
2049:Automated species identification
1566:10.1109/ICCICA52458.2021.9697287
579:Case Studies on Transport Policy
133:Shadow and drone movement effect
2034:Audio-visual speech recognition
1382:10.1109/ICCMC53470.2022.9753982
1337:10.1109/BigComp51126.2021.00032
897:"Diversity in Machine Learning"
784:10.1109/ITAIC49862.2020.9338958
332:Saran K B; Sreelekha G (2015).
170:Improvising existing techniques
1879:Recognition and categorization
1208:10.1109/ICIP46576.2022.9897990
377:Nemade, Bhushan (2016-01-01).
304:Use of UAVs in law enforcement
1:
2143:Optical character recognition
2074:Content-based image retrieval
239:Feature Pyramid Network (FPN)
61:(Traffic video Surveillance,
53:An example of object tracking
828:10.1109/ICNSURV.2016.7486437
299:Convolutional neural network
1656:10.1109/IGARSS.2010.5654227
1522:10.1109/ACCESS.2019.2956172
1427:10.1109/ICCTCT.2018.8550851
924:10.1109/ACCESS.2019.2917620
396:10.1016/j.procs.2016.03.052
89:Problems with small objects
2236:
2039:Automatic image annotation
1874:Noise reduction techniques
592:10.1016/j.cstp.2019.06.001
477:10.1007/s12652-022-03891-0
309:Underwater computer vision
2191:
2004:Free viewpoint television
383:Procedia Computer Science
346:10.1109/ICCC.2015.7432948
282:Vehicle re-identification
157:YOLOv5 and SAHI interface
2069:Computer-aided diagnosis
1136:10.1109/ICTAI.2019.00251
1070:10.1109/CVPRW.2019.00084
22:is a particular case of
2131:Moving object detection
2121:Medical image computing
1884:Research infrastructure
1854:Image sensor technology
432:10.1145/2733373.2806349
226:to define anchor size.
165:YOLOv7 detection output
149:YOLOv5 detection result
2168:Video content analysis
2136:Small object detection
1915:Computer stereo vision
1705:10.1109/TPAMI.2006.155
985:Cite journal requires
964:10.5281/zenodo.3908559
166:
158:
150:
134:
63:Small object retrieval
54:
20:Small object detection
2173:Video motion analysis
1984:Structure from motion
1930:3D object recognition
218:Auto learning anchors
164:
156:
148:
132:
79:Traffic flow analysis
71:Maritime surveillance
52:
2096:Foreground detection
2079:Reverse image search
2059:Bioimage informatics
2029:Activity recognition
1761:10.1155/2020/3738108
1650:. pp. 517–519.
1192:. pp. 966–970.
878:10.1155/2020/3189691
2163:Autonomous vehicles
2101:Gesture recognition
1964:2D to 3D conversion
1263:2018SPIE10615E..1EC
524:2022Senso..22.3862I
2178:Video surveillance
2116:Landmark detection
2024:3D pose estimation
2009:Volumetric capture
1969:Gaussian splatting
1925:Object recognition
1839:Commercial systems
1271:10.1117/12.2304811
274:Other applications
167:
159:
151:
135:
96:You Only Look Once
55:
2202:
2201:
2111:Image restoration
2054:Augmented reality
2019:
2018:
1999:4D reconstruction
1951:3D reconstruction
1844:Feature detection
1665:978-1-4244-9565-8
1620:978-1-7281-6486-1
1575:978-1-6654-2040-2
1515:: 172443–172469.
1481:978-1-6654-1244-5
1436:978-1-5386-3702-9
1391:978-1-6654-1028-1
1346:978-1-7281-8924-6
1217:978-1-6654-9620-9
1145:978-1-7281-3798-8
1079:978-1-7281-2506-0
1016:Google Developers
837:978-1-5090-2149-9
793:978-1-7281-5244-8
533:10.3390/s22103862
441:978-1-4503-3459-4
355:978-1-4673-7349-4
252:Add-on techniques
224:K-means algorithm
200:Data augmentation
67:Anomaly detection
50:
2227:
2126:Object detection
2091:Face recognition
1974:Shape from focus
1947:
1834:Digital geometry
1808:
1801:
1794:
1785:
1766:
1765:
1763:
1739:
1733:
1732:
1699:(8): 1319–1334.
1684:
1678:
1677:
1639:
1633:
1632:
1594:
1588:
1587:
1549:
1543:
1542:
1524:
1500:
1494:
1493:
1455:
1449:
1448:
1421:. pp. 1–5.
1410:
1404:
1403:
1365:
1359:
1358:
1320:
1314:
1313:
1311:
1299:
1293:
1292:
1256:
1236:
1230:
1229:
1201:
1185:
1179:
1178:
1176:
1164:
1158:
1157:
1119:
1113:
1112:
1110:
1098:
1092:
1091:
1053:
1047:
1046:
1044:
1032:
1026:
1025:
1023:
1022:
1008:
1002:
1001:
994:
988:
983:
981:
973:
971:
970:
951:
945:
944:
926:
916:
892:
883:
882:
880:
856:
850:
849:
812:
806:
805:
767:
761:
760:
758:
746:
740:
739:
737:
724:
718:
717:
715:
703:
697:
696:
694:
682:
676:
675:
673:
661:
655:
654:
652:
640:
634:
633:
631:
619:
613:
612:
594:
570:
564:
563:
553:
535:
503:
497:
496:
460:
454:
453:
415:
409:
408:
398:
374:
368:
367:
329:
285:Animal detection
259:Hyper-parameters
184:machine learning
121:) and software (
51:
32:state-of-the-art
24:object detection
2235:
2234:
2230:
2229:
2228:
2226:
2225:
2224:
2205:
2204:
2203:
2198:
2187:
2158:Robotic mapping
2106:Image denoising
2015:
1936:
1903:
1869:Motion analysis
1817:
1815:Computer vision
1812:
1774:
1769:
1741:
1740:
1736:
1686:
1685:
1681:
1666:
1641:
1640:
1636:
1621:
1596:
1595:
1591:
1576:
1551:
1550:
1546:
1502:
1501:
1497:
1482:
1457:
1456:
1452:
1437:
1412:
1411:
1407:
1392:
1367:
1366:
1362:
1347:
1322:
1321:
1317:
1301:
1300:
1296:
1281:
1238:
1237:
1233:
1218:
1187:
1186:
1182:
1166:
1165:
1161:
1146:
1121:
1120:
1116:
1100:
1099:
1095:
1080:
1055:
1054:
1050:
1034:
1033:
1029:
1020:
1018:
1010:
1009:
1005:
995:
984:
974:
968:
966:
953:
952:
948:
907:: 64323–64350.
894:
893:
886:
858:
857:
853:
838:
814:
813:
809:
794:
769:
768:
764:
748:
747:
743:
726:
725:
721:
705:
704:
700:
684:
683:
679:
663:
662:
658:
642:
641:
637:
621:
620:
616:
572:
571:
567:
505:
504:
500:
462:
461:
457:
442:
417:
416:
412:
376:
375:
371:
356:
331:
330:
326:
322:
295:
276:
267:
254:
241:
232:
220:
208:
193:
180:
172:
140:
91:
83:Object tracking
75:Drone surveying
42:
40:
17:
12:
11:
5:
2233:
2231:
2223:
2222:
2217:
2207:
2206:
2200:
2199:
2192:
2189:
2188:
2186:
2185:
2183:Video tracking
2180:
2175:
2170:
2165:
2160:
2155:
2153:Remote sensing
2150:
2145:
2140:
2139:
2138:
2133:
2123:
2118:
2113:
2108:
2103:
2098:
2093:
2088:
2083:
2082:
2081:
2071:
2066:
2064:Blob detection
2061:
2056:
2051:
2046:
2041:
2036:
2031:
2026:
2020:
2017:
2016:
2014:
2013:
2012:
2011:
2006:
1996:
1991:
1989:View synthesis
1986:
1981:
1976:
1971:
1966:
1961:
1955:
1953:
1944:
1938:
1937:
1935:
1934:
1933:
1932:
1922:
1920:Motion capture
1917:
1911:
1909:
1905:
1904:
1902:
1901:
1896:
1891:
1886:
1881:
1876:
1871:
1866:
1861:
1856:
1851:
1846:
1841:
1836:
1831:
1825:
1823:
1819:
1818:
1813:
1811:
1810:
1803:
1796:
1788:
1782:
1781:
1773:
1772:External links
1770:
1768:
1767:
1734:
1679:
1664:
1634:
1619:
1589:
1574:
1544:
1495:
1480:
1450:
1435:
1405:
1390:
1360:
1345:
1315:
1294:
1279:
1231:
1216:
1180:
1159:
1144:
1114:
1093:
1078:
1048:
1027:
1003:
987:|journal=
946:
884:
851:
836:
807:
792:
762:
741:
719:
698:
677:
656:
635:
614:
585:(1): 215–228.
565:
498:
455:
440:
410:
369:
354:
323:
321:
318:
317:
316:
311:
306:
301:
294:
291:
290:
289:
288:Fish detection
286:
283:
280:
279:Crowd counting
275:
272:
266:
263:
253:
250:
243:Use a feature
240:
237:
231:
228:
219:
216:
207:
204:
192:
189:
179:
176:
171:
168:
139:
136:
127:
126:
111:
103:
90:
87:
39:
36:
28:aerial imagery
15:
13:
10:
9:
6:
4:
3:
2:
2232:
2221:
2218:
2216:
2215:Image sensors
2213:
2212:
2210:
2197:
2196:
2195:Main category
2190:
2184:
2181:
2179:
2176:
2174:
2171:
2169:
2166:
2164:
2161:
2159:
2156:
2154:
2151:
2149:
2148:Pose tracking
2146:
2144:
2141:
2137:
2134:
2132:
2129:
2128:
2127:
2124:
2122:
2119:
2117:
2114:
2112:
2109:
2107:
2104:
2102:
2099:
2097:
2094:
2092:
2089:
2087:
2084:
2080:
2077:
2076:
2075:
2072:
2070:
2067:
2065:
2062:
2060:
2057:
2055:
2052:
2050:
2047:
2045:
2042:
2040:
2037:
2035:
2032:
2030:
2027:
2025:
2022:
2021:
2010:
2007:
2005:
2002:
2001:
2000:
1997:
1995:
1992:
1990:
1987:
1985:
1982:
1980:
1977:
1975:
1972:
1970:
1967:
1965:
1962:
1960:
1957:
1956:
1954:
1952:
1948:
1945:
1943:
1939:
1931:
1928:
1927:
1926:
1923:
1921:
1918:
1916:
1913:
1912:
1910:
1906:
1900:
1897:
1895:
1892:
1890:
1887:
1885:
1882:
1880:
1877:
1875:
1872:
1870:
1867:
1865:
1862:
1860:
1857:
1855:
1852:
1850:
1847:
1845:
1842:
1840:
1837:
1835:
1832:
1830:
1827:
1826:
1824:
1820:
1816:
1809:
1804:
1802:
1797:
1795:
1790:
1789:
1786:
1779:
1776:
1775:
1771:
1762:
1757:
1753:
1749:
1745:
1738:
1735:
1730:
1726:
1722:
1718:
1714:
1710:
1706:
1702:
1698:
1694:
1690:
1683:
1680:
1675:
1671:
1667:
1661:
1657:
1653:
1649:
1645:
1638:
1635:
1630:
1626:
1622:
1616:
1612:
1608:
1604:
1600:
1593:
1590:
1585:
1581:
1577:
1571:
1567:
1563:
1559:
1555:
1548:
1545:
1540:
1536:
1532:
1528:
1523:
1518:
1514:
1510:
1506:
1499:
1496:
1491:
1487:
1483:
1477:
1473:
1469:
1465:
1461:
1454:
1451:
1446:
1442:
1438:
1432:
1428:
1424:
1420:
1416:
1409:
1406:
1401:
1397:
1393:
1387:
1383:
1379:
1375:
1371:
1364:
1361:
1356:
1352:
1348:
1342:
1338:
1334:
1330:
1326:
1319:
1316:
1310:
1305:
1298:
1295:
1290:
1286:
1282:
1280:9781510617414
1276:
1272:
1268:
1264:
1260:
1255:
1250:
1246:
1242:
1235:
1232:
1227:
1223:
1219:
1213:
1209:
1205:
1200:
1195:
1191:
1184:
1181:
1175:
1170:
1163:
1160:
1155:
1151:
1147:
1141:
1137:
1133:
1129:
1125:
1118:
1115:
1109:
1104:
1097:
1094:
1089:
1085:
1081:
1075:
1071:
1067:
1063:
1059:
1052:
1049:
1043:
1038:
1031:
1028:
1017:
1013:
1007:
1004:
999:
992:
979:
965:
961:
957:
950:
947:
942:
938:
934:
930:
925:
920:
915:
910:
906:
902:
898:
891:
889:
885:
879:
874:
870:
866:
862:
855:
852:
847:
843:
839:
833:
829:
825:
821:
817:
811:
808:
803:
799:
795:
789:
785:
781:
777:
773:
766:
763:
757:
752:
745:
742:
736:
731:
723:
720:
714:
709:
702:
699:
693:
688:
681:
678:
672:
667:
660:
657:
651:
646:
639:
636:
630:
625:
618:
615:
610:
606:
602:
598:
593:
588:
584:
580:
576:
569:
566:
561:
557:
552:
547:
543:
539:
534:
529:
525:
521:
517:
513:
509:
502:
499:
494:
490:
486:
482:
478:
474:
470:
466:
459:
456:
451:
447:
443:
437:
433:
429:
425:
421:
414:
411:
406:
402:
397:
392:
388:
384:
380:
373:
370:
365:
361:
357:
351:
347:
343:
339:
335:
328:
325:
319:
315:
312:
310:
307:
305:
302:
300:
297:
296:
292:
287:
284:
281:
278:
277:
273:
271:
264:
262:
260:
251:
249:
246:
238:
236:
229:
227:
225:
217:
215:
213:
205:
203:
201:
197:
196:Deep learning
190:
188:
185:
177:
175:
169:
163:
155:
147:
143:
137:
131:
124:
120:
116:
112:
109:
104:
101:
97:
93:
92:
88:
86:
84:
80:
76:
72:
68:
64:
60:
37:
35:
33:
29:
25:
21:
2193:
2135:
2086:Eye tracking
1942:Applications
1908:Technologies
1894:Segmentation
1751:
1747:
1737:
1696:
1692:
1682:
1647:
1637:
1602:
1592:
1557:
1547:
1512:
1508:
1498:
1463:
1453:
1418:
1408:
1373:
1363:
1328:
1318:
1297:
1244:
1234:
1189:
1183:
1162:
1127:
1117:
1096:
1061:
1051:
1030:
1019:. Retrieved
1015:
1006:
978:cite journal
967:. Retrieved
949:
904:
900:
868:
864:
854:
819:
810:
775:
765:
744:
722:
701:
680:
659:
638:
617:
582:
578:
568:
518:(10): 3862.
515:
511:
501:
468:
458:
423:
413:
386:
382:
372:
337:
327:
268:
255:
242:
233:
221:
209:
194:
181:
173:
141:
113:At present,
59:surveillance
56:
19:
18:
1994:Visual hull
1889:Researchers
1509:IEEE Access
901:IEEE Access
389:: 402–409.
2209:Categories
1864:Morphology
1822:Categories
1309:2112.11798
1254:1709.05054
1199:2202.06934
1174:2003.07021
1108:1612.03144
1042:1812.00469
1021:2022-09-14
969:2022-09-14
914:1807.01477
756:2207.02696
735:2209.02976
713:2011.08036
692:2004.10934
671:1804.02767
650:1612.08242
629:1506.02640
320:References
212:resolution
123:algorithms
108:pedestrian
1713:0162-8828
1629:221086279
1584:246663727
1539:209319743
1531:2169-3536
1490:245707782
1400:248131806
1355:232236614
1226:246823962
1154:211211764
1088:198903617
941:206491718
933:2169-3536
802:231824327
609:195543435
601:2213-624X
542:1424-8220
493:248951090
485:1868-5145
405:1877-0509
1899:Software
1859:Learning
1849:Geometry
1829:Datasets
1778:VisDrone
1754:: 1–13.
1721:16886866
1674:16812504
1445:54438440
1289:20592770
871:: 1–18.
846:21388151
560:35632270
364:14779393
293:See also
100:features
2220:Imaging
1729:1699015
1259:Bibcode
551:9143895
520:Bibcode
512:Sensors
450:9041849
245:pyramid
138:Methods
119:sensors
1727:
1719:
1711:
1672:
1662:
1627:
1617:
1582:
1572:
1537:
1529:
1488:
1478:
1443:
1433:
1398:
1388:
1353:
1343:
1287:
1277:
1224:
1214:
1152:
1142:
1086:
1076:
939:
931:
844:
834:
800:
790:
607:
599:
558:
548:
540:
491:
483:
448:
438:
403:
362:
352:
115:drones
81:, and
1725:S2CID
1670:S2CID
1625:S2CID
1580:S2CID
1535:S2CID
1486:S2CID
1441:S2CID
1396:S2CID
1351:S2CID
1304:arXiv
1285:S2CID
1249:arXiv
1222:S2CID
1194:arXiv
1169:arXiv
1150:S2CID
1103:arXiv
1084:S2CID
1037:arXiv
937:S2CID
909:arXiv
842:S2CID
798:S2CID
751:arXiv
730:arXiv
708:arXiv
687:arXiv
666:arXiv
645:arXiv
624:arXiv
605:S2CID
489:S2CID
446:S2CID
360:S2CID
1752:2020
1717:PMID
1709:ISSN
1660:ISBN
1615:ISBN
1570:ISBN
1527:ISSN
1476:ISBN
1431:ISBN
1386:ISBN
1341:ISBN
1275:ISBN
1212:ISBN
1140:ISBN
1074:ISBN
998:link
991:help
929:ISSN
869:2020
832:ISBN
788:ISBN
597:ISSN
556:PMID
538:ISSN
481:ISSN
436:ISBN
401:ISSN
350:ISBN
182:The
38:Uses
1756:doi
1701:doi
1652:doi
1607:doi
1562:doi
1517:doi
1468:doi
1423:doi
1378:doi
1333:doi
1267:doi
1204:doi
1132:doi
1066:doi
960:doi
919:doi
873:doi
824:doi
780:doi
587:doi
546:PMC
528:doi
473:doi
428:doi
391:doi
342:doi
2211::
1750:.
1746:.
1723:.
1715:.
1707:.
1697:28
1695:.
1691:.
1668:.
1658:.
1646:.
1623:.
1613:.
1601:.
1578:.
1568:.
1556:.
1533:.
1525:.
1511:.
1507:.
1484:.
1474:.
1462:.
1439:.
1429:.
1417:.
1394:.
1384:.
1372:.
1349:.
1339:.
1327:.
1283:.
1273:.
1265:.
1257:.
1220:.
1210:.
1202:.
1148:.
1138:.
1126:.
1082:.
1072:.
1060:.
1014:.
982::
980:}}
976:{{
958:.
935:.
927:.
917:.
903:.
899:.
887:^
867:.
863:.
840:.
830:.
818:.
796:.
786:.
774:.
603:.
595:.
581:.
577:.
554:.
544:.
536:.
526:.
516:22
514:.
510:.
487:.
479:.
471:.
467:.
444:.
434:.
422:.
399:.
387:79
381:.
358:.
348:.
336:.
85:.
77:,
73:,
69:,
65:,
30:,
1807:e
1800:t
1793:v
1764:.
1758::
1731:.
1703::
1676:.
1654::
1631:.
1609::
1586:.
1564::
1541:.
1519::
1513:7
1492:.
1470::
1447:.
1425::
1402:.
1380::
1357:.
1335::
1312:.
1306::
1291:.
1269::
1261::
1251::
1228:.
1206::
1196::
1177:.
1171::
1156:.
1134::
1111:.
1105::
1090:.
1068::
1045:.
1039::
1024:.
1000:)
993:)
989:(
972:.
962::
943:.
921::
911::
905:7
881:.
875::
848:.
826::
804:.
782::
759:.
753::
738:.
732::
716:.
710::
695:.
689::
674:.
668::
653:.
647::
632:.
626::
611:.
589::
583:8
562:.
530::
522::
495:.
475::
452:.
430::
407:.
393::
366:.
344::
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
