Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,61 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
task_categories:
|
4 |
+
- object-detection
|
5 |
+
- image-segmentation
|
6 |
+
- image-classification
|
7 |
+
---
|
8 |
+
## Pascal VOC
|
9 |
+
#### Dataset Summary
|
10 |
+
The Pascal Visual Object Classes (VOC) dataset is a widely used benchmark in the field of computer vision. It is designed for object detection, image classification, semantic segmentation, and action classification tasks. The dataset provides a comprehensive set of annotated images covering 20 object classes, allowing researchers to evaluate and compare the performance of various algorithms.
|
11 |
+
**Note**: This dataset repository contains all editions of PASCAL-VOC, each file is identified with the year.
|
12 |
+
|
13 |
+
#### Dataset Structure
|
14 |
+
**Images:** The dataset contains 178k images.
|
15 |
+
**Annotations:** Annotations include object bounding boxes, object class labels, segmentation masks, and action labels.
|
16 |
+
**Classes:** 20 object classes: person, bicycle, car, motorbike, aeroplane, bus, train, boat, bird, cat, dog, horse, sheep, cow, elephant, bear, zebra, giraffe, and potted plant.
|
17 |
+
**Supported Tasks**
|
18 |
+
**Image Classification:** Assigning a label to an image from a fixed set of categories.
|
19 |
+
**Object Detection:** Identifying objects within an image and drawing bounding boxes around them.
|
20 |
+
**Semantic Segmentation:** Assigning a class label to each pixel in the image.
|
21 |
+
**Action Classification:** Identifying the action being performed in the image.
|
22 |
+
|
23 |
+
#### Applications
|
24 |
+
The Pascal VOC dataset is used for:
|
25 |
+
|
26 |
+
- Benchmarking and evaluating computer vision algorithms.
|
27 |
+
- Training models for image classification, object detection, and segmentation tasks.
|
28 |
+
|
29 |
+
#### Data Collection and Annotation
|
30 |
+
|
31 |
+
**Data Sources**
|
32 |
+
The images were collected from Flickr and other sources, ensuring a diverse and representative sample of real-world scenes.
|
33 |
+
|
34 |
+
A**nnotation Process**
|
35 |
+
Annotations were carried out by a team of human annotators. Each image is labeled with:
|
36 |
+
|
37 |
+
- Bounding boxes for object detection.
|
38 |
+
- Class labels for each object.
|
39 |
+
- Pixel-wise segmentation masks for semantic segmentation.
|
40 |
+
- Action labels indicating the action performed by the objects in the image.
|
41 |
+
|
42 |
+
#### License
|
43 |
+
The Pascal VOC dataset is released under the Creative Commons Attribution 2.5 License. Users are free to share, adapt, and use the dataset, provided appropriate credit is given.
|
44 |
+
|
45 |
+
#### Citation
|
46 |
+
If you use the Pascal VOC dataset in your research, please cite the following paper:
|
47 |
+
|
48 |
+
```
|
49 |
+
@article{Everingham10,
|
50 |
+
author = {Mark Everingham and
|
51 |
+
Luc Gool and
|
52 |
+
Christopher K. I. Williams and
|
53 |
+
John Winn and
|
54 |
+
Andrew Zisserman},
|
55 |
+
title = {The Pascal Visual Object Classes (VOC) Challenge},
|
56 |
+
journal = {International Journal of Computer Vision},
|
57 |
+
volume = {88},
|
58 |
+
number = {2},
|
59 |
+
year = {2010},
|
60 |
+
pages = {303-338},
|
61 |
+
}
|