TY - JOUR
T1 - Incidents1M
T2 - A Large-Scale Dataset of Images with Natural Disasters, Damage, and Incidents
AU - Weber, Ethan
AU - Papadopoulos, Dim P.
AU - Lapedriza, Agata
AU - Ofli, Ferda
AU - Imran, Muhammad
AU - Torralba, Antonio
N1 - Publisher Copyright:
© 1979-2012 IEEE.
PY - 2023/4/1
Y1 - 2023/4/1
N2 - Natural disasters, such as floods, tornadoes, or wildfires, are increasingly pervasive as the Earth undergoes global warming. It is difficult to predict when and where an incident will occur, so timely emergency response is critical to saving the lives of those endangered by destructive events. Fortunately, technology can play a role in these situations. Social media posts can be used as a low-latency data source to understand the progression and aftermath of a disaster, yet parsing this data is tedious without automated methods. Prior work has mostly focused on text-based filtering, yet image and video-based filtering remains largely unexplored. In this work, we present the Incidents1M Dataset, a large-scale multi-label dataset which contains 977,088 images, with 43 incident and 49 place categories. We provide details of the dataset construction, statistics and potential biases; introduce and train a model for incident detection; and perform image-filtering experiments on millions of images on Flickr and Twitter. We also present some applications on incident analysis to encourage and enable future work in computer vision for humanitarian aid. Code, data, and models are available at http://incidentsdataset.csail.mit.edu.
AB - Natural disasters, such as floods, tornadoes, or wildfires, are increasingly pervasive as the Earth undergoes global warming. It is difficult to predict when and where an incident will occur, so timely emergency response is critical to saving the lives of those endangered by destructive events. Fortunately, technology can play a role in these situations. Social media posts can be used as a low-latency data source to understand the progression and aftermath of a disaster, yet parsing this data is tedious without automated methods. Prior work has mostly focused on text-based filtering, yet image and video-based filtering remains largely unexplored. In this work, we present the Incidents1M Dataset, a large-scale multi-label dataset which contains 977,088 images, with 43 incident and 49 place categories. We provide details of the dataset construction, statistics and potential biases; introduce and train a model for incident detection; and perform image-filtering experiments on millions of images on Flickr and Twitter. We also present some applications on incident analysis to encourage and enable future work in computer vision for humanitarian aid. Code, data, and models are available at http://incidentsdataset.csail.mit.edu.
KW - Visual recognition
KW - disaster analysis
KW - image dataset
KW - incident detection
KW - scene understanding
KW - social media
UR - http://www.scopus.com/inward/record.url?scp=85135751725&partnerID=8YFLogxK
U2 - 10.1109/TPAMI.2022.3191996
DO - 10.1109/TPAMI.2022.3191996
M3 - Article
C2 - 35905065
AN - SCOPUS:85135751725
SN - 0162-8828
VL - 45
SP - 4768
EP - 4781
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
IS - 4
ER -