v2.5.0
Datasets ⭐
EXAMPLE CODE
Animal Pose Dataset
Load Animal Pose Dataset in Python fast. 5,517 keypoint-labeled animal data samples from 5 categories. Stream Animal Pose Dataset while training ML models.
Visualization of the Animal Pose Dataset on the Activeloop Platform

Animal Pose Dataset

What is Animal Pose Dataset?

The Animal Pose dataset contains 5,517 keypoint-labeled animal data samples from 5 categories scattered throughout 4000 photographs. After annotation, there are more than 20 keypoints. In addition, the dataset includes only-bounding-box annotations for further 7 animal types. A collection of animal poses may be utilized for a unique cross-domain adaption strategy to transfer animal pose knowledge from labeled animal classes to unlabeled animal classes.

Download Animal Pose Dataset in Python

Instead of downloading the Animal Pose dataset in Python, you can effortlessly load it in Python via our open-source package Hub with just one line of code.

Load Animal Pose Dataset with Keypoint in Python

1
import hub
2
ds = hub.load('hub://activeloop/animal-pose-keypoint-labeled')
Copied!

Load Animal Pose Dataset without Keypoint in Python

1
import hub
2
ds = hub.load('hub://activeloop/animal-pose-keypoint-unlabeled')
Copied!

Animal Pose Dataset Structure

Animal Pose Data Fields

For Pose Estimation with Keypoints
  • images: tensor containing the images.
  • box/boxes: tensor containing the bounding box coordinates.
  • box/supercategories: tensor containing the numerical label that represents the index of the supercategory in the list.
  • box/categories: tensor containing the numerical label that represents the index of the category in the list.
  • keypoint/keypoints: tensor containing the the keypoints.
  • keypoint/supercategories: tensor containing the numerical label that represents the index of the supercategory in the list.
  • keypoint/categories: tensor containing the numerical label that represents the index of the category in the list.
For Pose Estimation without Keypoints
  • images: tensor containing the images.
  • box/boxes: tensor containing the bounding box coordinates.
  • box/labels: tensor containing the numerical label that represents the index of the category in the list.

How to use Animal Pose Dataset with PyTorch and TensorFlow in Python

Train a model on Animal Pose dataset with PyTorch in Python

Let's use Hub's built-in PyTorch one-line dataloader to connect the data to the compute:
1
dataloader = ds.pytorch(num_workers=0, batch_size=4, shuffle=False)
Copied!

Train a model on Animal Pose Dataset with TensorFlow in Python

1
dataloader = ds.tensorflow()
Copied!

Animal Pose Dataset Creation

Source Data

The start point of the dataset is a collection of PASCAL 2011 keypoint annotations supplied by UC, Berkeley, to which additional annotations and photos were contributed. Some of the photos are taken from the Animals-10 dataset.

Additional Information about Animal Pose Dataset

Animal Pose Dataset Description

Animal Pose Dataset Curators

Jinkun Cao, Hongyang Tang, Hao-Shu Fang, Xiaoyong Shen, Cewu Lu, Yu-Wing Tai

Animal Pose Dataset Licensing Information

Hub users may have access to a variety of publicly available datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have a license to use the datasets. It is your responsibility to determine whether you have permission to use the datasets under their license. If you're a dataset owner and do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thank you for your contribution to the ML community!

Animal Pose Dataset Citation Information

1
@inproceedings{cao2019cross,
2
title={Cross-domain adaptation for animal pose estimation},
3
author={Cao, Jinkun and Tang, Hongyang and Fang, Hao-Shu and Shen, Xiaoyong and Lu, Cewu and Tai, Yu-Wing},
4
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
5
pages={9498--9507},
6
year={2019}
7
}
Copied!