PlatformMarketplaceSolutionsResourcesOpen DatasetsCommunityCompany
Finger Tapping Hand Keypoints Dataset
initial commit
Apr 11, 2022 2:34 AM


Subject 1Subject 2Subject 3Subject 4

This dataset contains 782 cropped images about the right hand's exercise performances corresponding to "finger appose" and "appose finger sucession" from 4 different subjects and respective annotations for them.

Appose Finger Sucession means the participant is instructed to tap each finger against the thumb in order of Thumb to Index, Middle, Ring, and Little finger respectively. This sequence is to be repeated as fast as possible. The participant is told not to tap in the backward order.

Finger Tapping means the participant is instructed to tap the index finger against the thumb as fast as possible.


Annotations for images are stored in CSV files. The columns correspond to these fields:

  1. Frame number: This is the frame number of each image from different subjects.
  2. translated_centroid: coordinates of the translated centroid of the right hand from the original cropped image.
  3. right_wrist: coordinates of the right wrist.
  4. right_thumb: coordinates of the right thumb tip.
  5. right_index: coordinates of the right index finger tip.
  6. right_middle: coordinates of the right middle finger tip.
  7. right_ring: coordinates of the right ring finger tip.
  8. right_little: coordinates of the right little finger tip.

Each coordinate has two columns, representing x and y respectively. Therefore the CSV file has 15 columns in total.


Image files are labelled in the format: "subject(#subject_number)_fingercount_cropframe_(#frame_num).jpg". Here (#subject_number) corresponds to subjects from 1 to 4 and (#frame_num) correspond to the image frame number.

The original images are of 640 X 480, while these cropped images collected in HKD are of 257 X 257 and are lossely cropped right hands from the participants' captured RGB frames.

The csv files are labelled in the format: "subject(#subject_nember)_fingercount_2D_Annotations_cropped.csv", encoded by 'utf-8'.

# Citation 
author = {Gattupalli, Srujana and Babu, Ashwin Ramesh and Brady, James Robert and Makedon, Fillia and Athitsos, Vassilis},
title = {Towards Deep Learning Based Hand Keypoints Detection for Rapid Sequential Movements from RGB Images},
year = {2018},
isbn = {9781450363907},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {},
doi = {10.1145/3197768.3201538},
abstract = {Hand keypoints detection and pose estimation has numerous applications in computer vision, but it is still an unsolved problem in many aspects. An application of hand keypoints detection is in performing cognitive assessments of a subject by observing the performance of that subject in physical tasks involving rapid finger motion. As a part of this work, we introduce a novel hand keypoints benchmark dataset that consists of hand gestures recorded specifically for cognitive behavior monitoring. We explore the state of the art methods in hand keypoint detection and we provide quantitative evaluations for the performance of these methods on our dataset. In future, these results and our dataset can serve as a useful benchmark for hand keypoint recognition for rapid finger movements.},
booktitle = {Proceedings of the 11th PErvasive Technologies Related to Assistive Environments Conference},
pages = {31–37},
numpages = {7},
keywords = {hand pose recognition, cognitive behavior assessment, convolutional neural networks, hand keypoints detection, computer vision, gesture recognition, deep learning},
location = {Corfu, Greece},
series = {PETRA '18}
Data Preview
List Dataset Files
🎉Many thanks to Hello Dataset for contributing the dataset
Basic Information
Application ScenariosPerson
TasksAction/Event DetectionPose Estimation
Updated on2022-04-11 02:56:04
Data TypeImage
Data Volume782
Annotation Amount782
File Size5.38MB
Copyright Owner
Srujana Gattupalli