mevadata.yaml

Name: Multiview Extended Video with Activities (MEVA)
Description: |
  The Multiview Extended Video with Activities (MEVA) dataset consists
  video data of human activity, both scripted and unscripted,
  collected with roughly 100 actors over several weeks.  The data was
  collected with 29 cameras with overlapping and non-overlapping
  fields of view. The current release consists of about 328 hours
  (516GB, 4259 clips) of video data, as well as 4.6 hours (26GB) of
  UAV data. Other data includes GPS tracks of actors, camera models,
  and a site map. We have also released annotations for roughly 184 hours of
  data. Further updates are planned.
Documentation: http://mevadata.org
Contact: mevadata@kitware.com
ManagedBy: "[Kitware](http://www.kitware.com/)"
UpdateFrequency: |
  We anticipate two or three updates as more
  video data is released to the public.
Tags:
  - aws-pds
  - computer vision
  - urban
  - us
  - video
License: http://mevadata.org/resources/MEVA-data-license.txt
Resources:
  - Description: AVI video clips and collection site map
    ARN: arn:aws:s3:::mevadata-public-01
    Region: us-east-1
    Type: S3 Bucket
DataAtWork:
  Publications:
    - Title: "MEVA: A Large-Scale Multiview, Multimodal Video Dataset for Activity Detection"
      URL: https://openaccess.thecvf.com/content/WACV2021/html/Corona_MEVA_A_Large-Scale_Multiview_Multimodal_Video_Dataset_for_Activity_Detection_WACV_2021_paper.html
      AuthorName: Kellie Corona, Katie Osterdahl, Roderic Collins, Anthony Hoogs
    - Title: "ActEV: Activities in Extended Video"
      URL: https://actev.nist.gov/
      AuthorName: National Institute of Standards and Technology (NIST)
    - Title: "TinyAction Challenge: Recognizing Real-world Low-resolution Activities in Videos"
      URL: https://arxiv.org/abs/2107.11494
      AuthorName: Praveen Tirupattur, Aayush J Rana, Tushar Sangam, Shruti Vyas, Yogesh S Rawat, Mubarak Shah