Search
Description
Images taken from bee hive with tagged bees. The images are random frames from movies recorded in may 2017 and 2018. All images are taken from full HD video stream.
The images filenames are the same as filename of source video, so it should be easy to match. Labels inside each file follow a name convention:
e.g. rawdata/DSCN9958_segment_007_frame_210_scale_0_8.bmp means that:
* video filename is DSCN9958
* segment number is 7 (it's 8th frame chosen from this video, numbers start from 0)
* frame number is 210 (it's 211th frame of all frames in this video, numbers start from 0)
* scale of tagged video is 0.8 ("_" instead of "."), to use labels for full scale videos, you need to unscale the labels
The tagging of the images was done manually using objectmaker.exe tool (supplied with the part 1 of the data, see references). The tags are stored in text files with names identical as the movie filename and has the following format: "image_name number_of_tags x_coordinate_of_top_lef_corner y_coordinate_of_top_lef_corner x_coordinate_of_bottom_right_corner y_coordinate_of_bottom_right_corner". The coordinate set is provided number_of_tags times.
Dataset file
hexmd5(md5(part1)+md5(part2)+...)-{parts_count} where a single part of the file is 512 MB in size.Example script for calculation:
https://github.com/antespi/s3md5
File details
- License:
-
open in new tab
CC BY-SAShare-alike
Details
- Year of publication:
- 2020
- Verification date:
- 2020-12-30
- Dataset language:
- English
- Fields of science:
-
- information and communication technology (Engineering and Technology)
- DOI:
- DOI ID 10.34808/66c8-j269 open in new tab
- Series:
- Verified by:
- Gdańsk University of Technology
Keywords
References
Cite as
Authors
seen 146 times