DiGyt commited on
Commit
6c1bfa2
1 Parent(s): c98f2b6

Update README.md

Browse files

Filled last missing fields

Files changed (1) hide show
  1. README.md +4 -3
README.md CHANGED
@@ -107,7 +107,7 @@ A total of 565 categories were selected based on the following: 1) their word fr
107
 
108
  ### Curation Rationale
109
 
110
- ...TODO
111
 
112
  ### Source Data
113
 
@@ -125,7 +125,7 @@ The dataset was tested to exclude sensitive images using Yahoo's Open NSFW detec
125
 
126
  ### Social Impact of Dataset
127
 
128
- ...TODO
129
 
130
  ### Discussion of Biases
131
 
@@ -162,4 +162,5 @@ Ecoset is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 2
162
 
163
  ### Contributions
164
 
165
- Thanks to #TODO
 
 
107
 
108
  ### Curation Rationale
109
 
110
+ More information on the curation of the dataset can be found in the [original publication](https://doi.org/10.1073/pnas.2011417118).
111
 
112
  ### Source Data
113
 
 
125
 
126
  ### Social Impact of Dataset
127
 
128
+ Large-scale image-label datasets such as ImageNet provide the backbone of modern Computer Vision. However, such large datasets often suffer from problems like mislabeling, category biases, misrepresentations, and unsafe content. Ecoset was created with the aim to reduce these biases and consequently improve the social impact of Computer Vision techniques trained on the dataset. More information on the social impact of the dataset can be found in the [original publication](https://doi.org/10.1073/pnas.2011417118).
129
 
130
  ### Discussion of Biases
131
 
 
162
 
163
  ### Contributions
164
 
165
+ The ecoset dataloader and dataset card was created by @DiGyt on behalf of @kietzmannlab.
166
+ For questions and suggestions feel free to reach out.