Tip:
Highlight text to annotate it
X
The 50 Salads dataset captures
25 people preparing 2 salads each.
Participants cover a wide range of
food preparation experience,
age, gender,
and ethnic background.
Task orderings were randomized,
inducing strong temporal variation.
The 50 Salads dataset contains:
1. RGBD-video data from a downward-facing camera.
2. Data from accelerometers attached to 7 kitchen objects.
3. Annotations of pre-, core- and post-phases of individudal steps in the recipe.
Usecases include:
activity recognition, activity spotting,
sequence analysis, progress tracking,
sensor fusion, transfer learning,
skill assessment,
and user-adaptation.
The 50 Salads dataset is intended for research purposes only.
As such it cannot be used commercially.
Reference must be made to the following publication
whenever research making use of this dataset
is reported in any academic publication or research report.
Sebastian Stein and Stephen J. McKenna
"Combining Embedded Accelerometers with Computer Vision
for Recognizing Food Preparation Activities"
In Proceedings of UbiComp 2013