Update README.md
Browse files
README.md
CHANGED
|
@@ -23,8 +23,11 @@ Content: The dataset features:
|
|
| 23 |
* Complex annotations with up to 500+ labels per frame in some cases
|
| 24 |
|
| 25 |
Sources: Frames were extracted from in-store video footage recorded in cosmetic retail environments (e.g., stores like Nocibé or Sephora, though not explicitly named).
|
|
|
|
| 26 |
Format: Raw images and their respective annotations, in XML format are available. Dataset was developed using CVAT and organized into subsets to optimize performance during extraction and analysis.
|
|
|
|
| 27 |
Annotation tool used: CVAT
|
|
|
|
| 28 |
Annotations: Annotations include bounding boxes / polygons for individual products and shelf layouts. For projects requiring additional metadata, feel free to contact us.
|
| 29 |
|
| 30 |
|
|
|
|
| 23 |
* Complex annotations with up to 500+ labels per frame in some cases
|
| 24 |
|
| 25 |
Sources: Frames were extracted from in-store video footage recorded in cosmetic retail environments (e.g., stores like Nocibé or Sephora, though not explicitly named).
|
| 26 |
+
|
| 27 |
Format: Raw images and their respective annotations, in XML format are available. Dataset was developed using CVAT and organized into subsets to optimize performance during extraction and analysis.
|
| 28 |
+
|
| 29 |
Annotation tool used: CVAT
|
| 30 |
+
|
| 31 |
Annotations: Annotations include bounding boxes / polygons for individual products and shelf layouts. For projects requiring additional metadata, feel free to contact us.
|
| 32 |
|
| 33 |
|