Pascal VOC

  Dataset Summary

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. Note: This dataset repository contains all editions of… See the full description on the dataset page: https://huggingface.co/datasets/merve/pascal-voc.

The Pascal VOC 2007 dataset is a dataset of images with object annotations. The dataset consists of 20 object classes, and each image is labeled with a bounding box for each object. The Pascal VOC 2007 dataset has been used to train and evaluate a variety of object detection algorithms.

This dataset contains the data from the PASCAL Visual Object Classes Challenge, corresponding to the Classification and Detection competitions.

In the Classification competition, the goal is to predict the set of labels contained in the image, while in the Detection competition the goal is to predict the bounding box and label of each individual object. WARNING: As per the official dataset, the test set of VOC2012 does not contain annotations.

To use this dataset:

import tensorflow_datasets as tfds

ds = tfds.load('voc', split='train')
for ex in ds.take(4):
 print(ex)

See the guide for more informations on tensorflow_datasets.

https://storage.googleapis.com/tfds-data/visualization/fig/voc-2007-5.0.0.png" alt="Visualization" width="500px">

Pascal VOC 2012 is common benchmark for object detection. It contains common objects that one might find in images on the web.

https://i.imgur.com/y2sB9fD.png" alt="Image example">

Note: the test set is witheld, as is common with benchmark datasets.

You can think of it sort of like a baby COCO.

http://host.robots.ox.ac.uk/pascal/VOC/voc2007/index.html

20 classes: Person: person Animal: bird, cat, cow, dog, horse, sheep Vehicle: aeroplane, bicycle, boat, bus, car, motorbike, train Indoor: bottle, chair, dining table, potted plant, sofa, tv/monitor

Train/validation/test: 9,963 images containing 24,640 annotated objects.

VOC PASCAL 2007

## Overview

VOC PASCAL 2007 is a dataset for object detection tasks - it contains VOC PASCAL 2007 annotations for 4,952 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

The Semantic PASCAL-Part dataset

The Semantic PASCAL-Part dataset is the RDF version of the famous PASCAL-Part dataset used for object detection in Computer Vision. Each image is annotated with bounding boxes containing a single object. Couples of bounding boxes are annotated with the part-whole relationship. For example, the bounding box of a car has the part-whole annotation with the bounding boxes of its wheels.

This original release joins Computer Vision with Semantic Web as the objects in the dataset are aligned with concepts from:

• the provided supporting ontology;
• the WordNet database through its synstes;
• the Yago ontology.

The provided Python 3 code (see the GitHub repo) is able to browse the dataset and convert it in RDF knowledge graph format. This new format easily allows the fostering of research in both Semantic Web and Machine Learning fields.

Structure of the semantic PASCAL-Part Dataset

This is the folder structure of the dataset:

• semanticPascalPart: it contains the refined images and annotations (e.g., small specific parts are merged into bigger parts) of the PASCAL-Part dataset in Pascal-voc style.
  • Annotations_set: the test set annotations in .xml format. For further information See the PASCAL VOC format here.
  • Annotations_trainval: the train and validation set annotations in .xml format. For further information See the PASCAL VOC format here.
  • JPEGImages_test: the test set images in .jpg format.
  • JPEGImages_trainval: the train and validation set images in .jpg format.
  • test.txt: the 2416 image filenames in the test set.
  • trainval.txt: the 7687 image filenames in the train and validation set.

The PASCAL-Part Ontology

The PASCAL-Part OWL ontology formalizes, through logical axioms, the part-of relationship between whole objects (22 classes) and their parts (39 classes). The ontology contains 85 logical axiomns in Description Logic in (for example) the following form:

Every potted_plant has exactly 1 plant AND
          has exactly 1 pot

We provide two versions of the ontology: with and without cardinality constraints in order to allow users to experiment with or without them. The WordNet alignment is encoded in the ontology as annotations. We further provide the WordNet_Yago_alignment.csv file with both WordNet and Yago alignments.

The ontology can be browsed with many Semantic Web tools such as:

• Protégé: a graphical tool for ongology modelling;
• OWLAPI: Java API for manipulating OWL ontologies;
• rdflib: Python API for working with the RDF format.
• RDF stores: databases for storing and semantically retrieve RDF triples. See here for some examples.

Citing semantic PASCAL-Part

If you use semantic PASCAL-Part in your research, please use the following BibTeX entry

@article{DBLP:journals/ia/DonadelloS16,
 author  = {Ivan Donadello and
        Luciano Serafini},
 title   = {Integration of numeric and symbolic information for semantic image
        interpretation},
 journal  = {Intelligenza Artificiale},
 volume  = {10},
 number  = {1},
 pages   = {33--47},
 year   = {2016}
}

The PASCAL VOC project:

Provides standardized image data sets for object class recognition Provides a common set of tools for accessing the data sets and annotations Enables evaluation and comparison of different methods Ran challenges evaluating performance on object class recognition (from 2005-2012, now finished)

Content

This dataset has the images of objects present in pascal Voc 2012-object detection dataset, grouped into directories according to their shapes. It also includes some negative samples extracted through ROI proposals. The images are organized in tfrecords format for training in TPUs.

Acknowledgement

This dataset is created by extracting objects from https://www.kaggle.com/huanghanchina/pascal-voc-2012. Thanks to Hans .

Pascal VOC 2012 Person Only

## Overview

Pascal VOC 2012 Person Only is a dataset for object detection tasks - it contains Person JdTc annotations for 9,006 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Wheat Pascal VOC

## Overview

Wheat Pascal VOC is a dataset for object detection tasks - it contains Germinant Gibberellic Mildew annotations for 1,746 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset is a set of additional annotations for PASCAL VOC 2010. It goes beyond the original PASCAL semantic segmentation task by providing annotations for the whole scene. The statistics section has a full list of 400+ labels. Since the dataset is an annotation of PASCAL VOC 2010, it has the same statistics as those of the original dataset. Training and validation contains 10,103 images while testing contains 9,637 images.

A benchmark dataset for visual object classes, containing 20 object classes and over 100,000 images.

Standardised image data sets for object class recognition - both 2007 and 2012 versions are provided here. The 2012 version has 20 classes. The train/val data has 11,530 images containing 27,450 ROI annotated objects and 6,929 segmentations.

Dataset

This dataset was created by Aladdin Persson

Contents

It contains the following files:

The development of robotic solutions for agriculture requires advanced perception capabilities that can work reliably in any crop stage. ROBOCARE is a project from CRIIS - INESC TEC, which focus on the development of robotics solutions for tomatoes' monitoring and harvesting inside greenhouses. Therefore, this project requires a reliable computer vision system to classify and localise tomatoes in the tomatoes' plants in the greenhouses. The current dataset was collected using a stereo camera (ZED camera) assembled in a static manipulator over a robotic platform to increase the similarity with the acquired data by the robot during its operation. The dataset was manually labeled using a dedicated annotation tool and is distributed under its full-size images (1280x720 px) or splitted images with 300x300 px considering an overlapping of 20% between adjacent images. This dataset is part of the journal paper submitted to Sensors and is distributted to allow the reproducibility of the performed work.

Context

Pascal VOC Dataset 2012, is the standard dataset for Image Segmentation, Detection, Localization, etc. In image segmentation: We need to predict per pixel prediction. Object Detection: We need to specify what classes are present in the given image. We can also bound them using a bounding box.

Content

It contains two directories one contains validation and training set and the other contains the test set. Inside the train_val directory, it contains an Image set which contains a text file that represents training and validation instances. For Every image, it provides Class labels and Objects Labels along with annotations. Labeled images contain the class label per pixel.

The same goes for the Test set. The predicted labels of the test set are also present inside SegmentationClass or Segmentation object depending on which application you are working on.

Acknowledgements

I downloaded the dataset from the standard PASCAL VOC site.

Inspiration

I made this dataset available so that everyone can use it to train their model for various different applications, and have a chance to embrace their knowledge in the respective field.

A benchmark for object detection

Pascal Voc

## Overview

Pascal Voc is a dataset for object detection tasks - it contains Person Tv Bottle Chair annotations for 7,368 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

The PASCAL-Part 2010 dataset is a set of additional annotations for PASCAL VOC 2010, providing segmentation masks for each body part of the object and silhouette annotation for categories that do not have a consistent set of parts.