HICO Classification Dataset

Estimated reading: 4 minutes

Visualization of the HICO Classification train dataset in the Deep Lake UI.

HICO Classification Dataset

What is HICO Classification Dataset?

The HICO (Humans Interacting with Common Objects) Classification dataset is a new benchmark for identifying human-object interactions (HOI). The important features of this dataset are a wide range of interactions with common object categories, a list of well-defined, sense-based HOI categories, and detailed labeling of cooccurring interactions with an object category in each image.

Download HICO Classification Dataset in Python

Instead of downloading the HICO classification dataset in Python, you can effortlessly load it in Python via our Deep Lake open-source with just one line of code.

Load HICO Classification Dataset Training Subset in Python

				
					import deeplake
ds = deeplake.load("hub://activeloop/hico-classification-train")

Load HICO Classification Dataset Testing Subset in Python

				
					import deeplake
ds = deeplake.load("hub://activeloop/hico-classification-test")

HICO Classification Dataset Structure

HICO Classification Data Fields

images: tensor containing the images.
actions: tensor containing labels that represent human actions. If there is no interaction with the object, it is represented as ‘no_interaction’
objects: tensor containing the labels that represent the objects that humans interact with.
action_definitions: tensor containing the meaning of the human actions.
action_synonyms: tensor containing the synonyms of the human actions.
action_ings: tensor containing the present continuous tense of the human actions.

HICO Classification Data Splits

The HICO Classification dataset training set is composed of 38109 images. Some of the images were corrupted and removed.
The HICO Classification dataset test set was composed of 9658 images. Some of the images were corrupted and removed.

How to use HICO Classification Dataset with PyTorch and TensorFlow in Python

Train a model on the HICO Classification dataset with PyTorch in Python

Let’s use Deep Lake built-in PyTorch one-line dataloader to connect the data to the compute:

				
					dataloader = ds.pytorch(num_workers=0, batch_size=4, shuffle=False)

Train a model on the HICO Classification dataset with TensorFlow in Python

				
					dataloader = ds.tensorflow()

HICO Classification Dataset Creation

Source Data

Data Collection and Normalization Information

The 600 HOI categories were constructed by defining a set of 80 common objects and for each object their respective common interactions. The common objects used were selected based on children’s vocabularies. By combining manually filtered results from MS-COCO and Google N-Gram, HOI obtained a set of “common” verbs for each object category.

Additional Information about HICO Classification Dataset

HICO Classification Dataset Description

Homepage:http://websites.umich.edu/~ywchao/hico/
Paper: www-personal.umich.edu/~ywchao/publications/chao_iccv2015.pdf
Point of Contact: [email protected]

HICO Classification Dataset Curators

Yu-Wei Chao, Zhan Wang, Yugeng He, Jiaxuan Wang, and Jia Deng

HICO Classification Dataset Licensing Information

Deep Lake 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!

HICO Classification Dataset Licensing Information

				
					@INPROCEEDINGS{chao:iccv2015,
  author = {Yu-Wei Chao and Zhan Wang and Yugeng He and Jiaxuan Wang and Jia Deng},
  title = {{HICO}: A Benchmark for Recognizing Human-Object Interactions in Images},
  booktitle = {Proceedings of the IEEE International Conference on Computer Vision},
  year = {2015},
}

HICO Classification Dataset FAQs

What is the HICO Classification dataset for Python?

HICO is an industry standard for detecting human-object interactions (HOI). Dataset features include a wide range of interactions with various object types. The HOI dataset has 600 categories that were constructed by defining a set of 80 common objects and their respective common interactions for each object.

How to download the HICO Classification dataset in Python?

You can load the HICO dataset fast with one line of code using the open-source package Activeloop Deep Lake in Python. See detailed instructions on how to load the HICO Classification Dataset training subset or the HICO Classification Dataset testing subset in Python.

How can I use HICO Classification dataset in PyTorch or TensorFlow?

You can stream the HICO Classification dataset while training a model in PyTorch or TensorFlow with one line of code using the open-source package Activeloop Deep Lake in Python. See detailed instructions on how to train a model on the HICO Classification dataset with PyTorch in Python or train a model on the HICO Classification dataset with TensorFlow in Python.