Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -154,9 +154,13 @@ size_categories:
|
|
154 |
"Wake Vision" is a large, high-quality dataset featuring over 6 million images, significantly exceeding the scale and diversity of
|
155 |
current tinyML datasets (100x). This dataset includes images with annotations of whether each image contains a person. Additionally,
|
156 |
it incorporates a comprehensive fine-grained benchmark to assess fairness and robustness, covering perceived gender, perceived age,
|
157 |
-
subject distance, lighting conditions, and depictions.
|
158 |
-
|
159 |
-
|
|
|
|
|
|
|
|
|
160 |
|
161 |
- **License:** [CC-BY 4.0]
|
162 |
|
|
|
154 |
"Wake Vision" is a large, high-quality dataset featuring over 6 million images, significantly exceeding the scale and diversity of
|
155 |
current tinyML datasets (100x). This dataset includes images with annotations of whether each image contains a person. Additionally,
|
156 |
it incorporates a comprehensive fine-grained benchmark to assess fairness and robustness, covering perceived gender, perceived age,
|
157 |
+
subject distance, lighting conditions, and depictions.
|
158 |
+
|
159 |
+
|
160 |
+
The Wake Vision labels are derrived from Open Image's annotations which are licensed by Google LLC under CC BY 4.0 license.
|
161 |
+
The images are listed as having a CC BY 2.0 license. Note from Open Images: "while we tried to identify images that are licensed
|
162 |
+
under a Creative Commons Attribution license, we make no representations or warranties regarding the license status of each
|
163 |
+
image and you should verify the license for each image yourself."
|
164 |
|
165 |
- **License:** [CC-BY 4.0]
|
166 |
|