\nLength: 30 minutes (37,423 frames)
\nResolution: 1920*480 pixels (stitched from three cameras).
\nThermal sensors: AXIS Q1922, uncooled micro bolometer 8-14 \u03bcm, 10mm focal length.
\nROI: People inside the red line illustrated below are counted. Corner coordinates: (26,306) (485,81) (656,41) (1292,60) (1497,110) (1902,295) (1392,479) (359,479).
\nAnnotation: Can be downloaded here<\/a>. The format is [timestamp] [number of people inside ROI].<\/p>\n
![\"Annotated](\"https:\/\/bbh.create.aau.dk\/wp-content\/uploads\/2012\/05\/thermal.png\" "\\"Thermal")
<\/p>\n
Thermal dataset 1<\/h3>\n
Person annotation dataset<\/p>\n
This dataset contains a\u00a0~16 minutes of video\u00a0(30,000 frames) recorded in a sports arena. Every person inside the court area is annotated.<\/p>\n
The Image below shows the camera placement and the visible part of the court.<\/p>\n
![\"\"](\"https:\/\/bbh.create.aau.dk\/wp-content\/uploads\/2012\/05\/cameraplacement1.png\" "\\"cameraplacement\\"")
<\/p>\n
The image below show an example of an annotated frame. The red dots mark the persons position on the court and the green line indicates the border of the court.<\/p>\n
![\"Annotated](\"https:\/\/bbh.create.aau.dk\/wp-content\/uploads\/2012\/05\/V9-20F34873.png\" "\\"Annotatedimage\\"")
<\/p>\n
The dataset consist of raw images grapped from the video and text files containing the image coordinates of the red dots and the total number of persons on the court. An exampel from a text file\u00a0can be seen below. The image coordinates are written x,y,x,y,x,y,x,y…<\/p>\n
framenumber: 26006;<\/pre>\nmanual image coordinates: 166,195,193,82,189,70,227,69,264,76,281,62,293,62,315,73;<\/pre>\nmanual number of persons: 8;<\/pre>\nframenumber: 26007;<\/pre>\nmanual image coordinates: 168,195,192,81,224,69,189,68,281,66,293,64,316,74;<\/pre>\nmanual number of persons: 7;<\/pre>\nWith the text files comes also a file named court.txt<\/em> that contains the corners for the court in image coordinates and a file named homography.txt<\/em> that contains a\u00a0homopgraphy to calculate the world coordinates of the persons. (0,0) is in the lower left corner of the court seen in the first picture.<\/p>\n
The original images can be downloaded here<\/a>.<\/p>\n
The text files can be downloaded here<\/a>.<\/p>\n
The annotated images can be downloaded here<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"
This page will contain thermal datasets available for download. For more information about the datasets contact rg [at] create.aau.dk Thermal dataset 2 People counting dataset Please cite: R. Gade, A. J\u00f8rgensen and T.B. Moeslund: Long-term Occupancy Analysis using Graph-Based Optimisation in Thermal Imagery. IEEE Conference on Computer Vision and Pattern Recognition 2013. Download images (10 … Continue reading “Dataset”<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-489","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=\/wp\/v2\/pages\/489","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=489"}],"version-history":[{"count":0,"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=\/wp\/v2\/pages\/489\/revisions"}],"wp:attachment":[{"href":"https:\/\/bbh.create.aau.dk\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=489"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}