Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
139
datasets available to search
ShareScore release 0.7.1
Dataset results
139 results for “gaze”
Data Artifact for "Gaze into the Pattern: Characterizing Spatial Patterns with Internal Temporal Correlations for Hardware Prefetching"
<p>This dataset contains the Ligra, PARSEC, GAP, and QMM traces used in our paper "Gaze into the Pattern: Characterizing Spatial Patterns with Internal Temporal Correlations for Hardware Prefetching", which is accepted by HPCA'25. These traces are provided as part of the AE proces.</p>
RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments
<p><strong>License + Attribution</strong></p> <p>This dataset is licensed under <a href="https://creativecommons.org/licenses/by-nc-sa/4.0/">CC BY-NC-SA 4.0</a>. Commercial usage is not permitted. If you use this dataset or the code in a scientific publication, please cite the following <a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Tobias_Fischer_RT-GENE_Real-Time_Eye_ECCV_2018_paper.html">paper</a>:</p> <blockquote> <p>@inproceedings{FischerECCV2018,<br> author = {Tobias Fischer and Hyung Jin Chang and Yiannis Demiris},<br> title = "{RT-GENE: Real-Time Eye Gaze Estimation in Natural Environments}",<br> booktitle = {European Conference on Computer Vision},<br> year = {2018},<br> month = {September},<br> pages = {339--357}<br> }</p> </blockquote> <p>This work was supported in part by the Samsung Global Research Outreach program, and in part by the EU Horizon 2020 Project PAL (643783-RIA).</p> <p>More information can be found on the Personal Robotic Lab's website: <a href="https://www.imperial.ac.uk/personal-robotics/software/">https://www.imperial.ac.uk/personal-robotics/software/</a>.</p> <p><strong>Overview</strong></p> <p>The dataset consists of two parts: 1) One where the eyetracking glasses were worn (and thus ground truth labels for head-pose and eye gaze are available; suffix <em>_glasses</em>), and 2) One with natural appearances (no eyetracking glasses are worn; suffix <em>_noglasses</em>). The <em>_noglasses</em> images were used to train subject-specific GANs, and these GANs were used to inpaint the region covered by the eyetracking glasses in the <em>_glasses</em> images.</p> <p>There is code accompanying this dataset: <a href="https://github.com/Tobias-Fischer/rt_gene">https://github.com/Tobias-Fischer/rt_gene</a>. Please use the issue tracker in the code respository if you have questions regarding the dataset.</p> <p><strong>Subjects / 3-Fold evaluation</strong></p> <p>15 participants were recorded in 17 sessions. Session 014 is a second recording of participant 002, and session 015 is a second recording of participant 005 (different days and different camera poses were used).</p> <p>We used a 3-fold evaluation, with the three folds consisting of the following sessions (test on one of the groups, training with the remaining two groups):</p> <ol> <li>'s001', 's002', 's008', 's010'</li> <li>'s003', 's004', 's007', 's009'</li> <li>'s005', 's006', 's011', 's012', 's013'</li> </ol> <p>The validation set consists of sessions 's014', 's015' and 's016'.</p> <p>While the MATLAB script (<em>prepare_dataset.m</em>; see code repository) creates train and test images for each subject, all images were used for the evaluation (see <em>evaluate_model.py</em>).</p> <p><strong>Labeled dataset (sXYZ_glasses)</strong></p> <p>The file for each subject contains the following information:</p> <ul> <li>label_combined.txt This is the main file containing labels. The formatting is as follows:<br> seq_number, [head pose: right(pos) / left(neg), up (pos) / down(neg)], [gaze: right(pos) / left(neg), up(pos) / down(neg)], timestamp</li> <li>label_headpose.txt This file contains more detail about the head pose of the subject.<br> seq_number, [head pose translation: further(pos) / closer(neg), left(pos) / right(neg), up(pos) / down(neg)], [head pose rotation: roll right(pos) / roll left(neg), down(pos) / up(neg), rotate left(pos), rotate right(neg)], timestamp</li> <li>kinect2_calibration.yaml<br> The kinect2_calibration.yaml file contains the camera projection matrix in ROS format (this file should not be required).</li> <li>kinect2_pose.txt<br> The kinect2_pose.txt file contains the pose of the Kinect with respect to the motion capture system (this file should not be required).</li> <li>"original" folder <ul> <li>The face_before_inpainting folder contains the face with a large margin to the left and right.</li> <li>The mask folder contains images indicating the regions of the eyetracking glasses, aligned with the images in the face_before_inpainting folder.</li> <li>The overlay folder contains images where the mask was overlaid on the face_before_inpainting images.</li> <li>The face folder contains the face image extracted using MTCNN with a tighter margin.</li> <li>The left and right folders contain the left and right eye image areas.</li> </ul> </li> <li>The face, left and right images were used as baseline comparison in the paper (Fig. 7 without inpainting).</li> <li>"inpainted" folder <ul> <li>The face_after_inpainting folder contains images corresponding to the ones in the face_before_inpainting folder after applying the inpainting.</li> <li>Then, the images contained in the face, left and right folders were extracted using MTCNN as above.</li> </ul> </li> </ul> <p><strong>Unlabeled dataset (sXYZ_noglasses)</strong></p> <ul> <li>kinect2_calibration.yaml<br> This file contains the camera projection matrix in ROS format (this file should not be required).</li> <li>kinect2_pose.txt<br> This file contains the pose of the Kinect with respect to the motion capture system (this file should not be required).</li> <li>"face" folder<br> This folder contains the faces that can be used to train the GANs (without eyetracking glasses being worn).</li> </ul>
Supplementary Movies (Humans use Predictive Gaze Strategies to Target Waypoints During Steering)
<p>Supplementary movies for the article Humans use Predictive Gaze Strategies to Target Waypoints During Steering</p>
statistical analysis and data: How personality shapes gaze behavior without compromising subtle emotion recognition.
<pre>CODE:<br>script_gca_X.R: scripts used for the growth curve analysis fits<br>script_fits_brms_no_tb.r: script containing the fits (except those related to the crowth curve analysis).<br>inpact_script_contrasts.r: contrasts of every fits (call inpact_script_plot.r and inpact_script_save.r)<br>inpact_script_plot.r: plots<br>inpact_script_save: save contrasts to csv and xlsx files<br><br>DATA:<br>cluster_df.Rda: personality <br>neutral.Rda: neutral trials<br>sdtg.Rda: signal detection theory parameters<br>resp_emo.Rda : raw data to emotional trials<br>df_et4_propn.Rda: eye tracking data, first exposure phase of the emotional trials (0-1000ms)<br>df_et4_prop.Rda: eye tracking data, second exposure phase of the emotional trials (1000-2000ms)<br>df_n4_propn.Rda: eye tracking data, first exposure phase of the neutral trials (0-1000ms)<br>df_n4_prop.Rda: eye tracking data, second exposure phase of the neutral trials (1000-2000ms)<br><br>gca_e_contrast.Rda: contrasts of the growth curve analysis fit for the eye Area of Interest (AOI)<br>gca_n_contrast.Rda: contrasts of the growth curve analysis fit for the nose AOI<br>gca_m_contrast.Rda: contrasts of the growth curve analysis fit for the mouth AOI<br><br>X.rds: fits<br><br><br>STIMULI:<br>- videos of the neutral and emotional facial expressions<br>- backward masks</pre>
Colorectal-polyps-gaze-dataset
<p><strong> </strong><strong>Gaze-based attention network </strong></p> <p>Automatic and accurate classification of colorectal polyps based on convolutional neural networks (CNNs) during endoscopy is vital for assisting endoscopists in diagnosis and treatment. However, this task remains challenging due to the difficult data acquisition and annotation process, the poor interpretability, and the weak clinical acceptance of the CNN models. To tackle these dilemmas, we propose an innovative approach that utilizes endoscopists' gaze attention information as an auxiliary supervisory signal to train a CNN-based model for colorectal polyps classification. Specifically, the endoscopists’ gaze information when reading endoscopic images is first recorded through an eye-tracker. Then, the gaze information is processed and applied to supervise the CNN model's attention via an attention consistency module. The proposed gaze-based attention network contains a classification module and an attention consistency module.</p> <p><strong>Colorectal-polyps-gaze-dataset</strong></p> <p>We constructed the colorectal polyps gaze dataset, which contained NBI images of colorectal polyps and the corresponding gaze attention images (i.e., gaze attention maps and gaze attention heatmaps). All data collection and annotation processes are carried out by following the tenets of the Declaration of Helsinki. Firstly, a junior endoscopist retrospectively searched the NBI observation data of colonoscopy at Xiangyang Central Hospital from January 1, 2023, to March 10, 2024. The criterion for inclusion is that patients with colorectal polyps must have corresponding pathology reports, that is to say, the pathological examination is the gold standard of the diagnosis. According to this criterion, we collected 585 NBI images with colorectal polyps of 87 patients. Secondly, a senior endoscopist classified the 585 NBI images into three categories based on the NICE classification method. During this classification process, the endoscopist’s eye movement information would be recorded and used to generate the gaze attention images. Finally, a patient-level data splitting was performed to develop and validate the proposed methods. The selected patients and their corresponding images were randomly split into three sets, of which 60% was utilized for training, 20% was used for validation and the rest was utilized for testing. In the training and validation sets, we had the original image and the gaze attention images which generated from eye movement information. In the test set, only the original NBI images were included to test the actual performance of the trained models. </p> <p> </p>
Assessing the impact of central and peripheral obstructions on visual behavior: insights from gaze-contingent eye-tracking studies
<p>This dataset is a collection of image stimuli, as well as gaze tracks collected using visual field masks, such as central and peripheral scotoma, in order to assess the way people with visual field loss process visual stimuli in digital environments.</p> <h3><strong>This dataset is structured as follows :</strong></h3> <ul> <li>Stimuli: 125 fullHD images (1920x1080) used during the experiment trials; 4 fullHD images used during the training phase.</li> <li>Trials_metadata: <ul> <li>Psychopy_outputs: various callbacks and logs files from the Psychopy experiment</li> <li>trials_metadata: .csv files containing metadata such as stimuli order and timestamps.</li> </ul> </li> <li>Raw_eyetracking_outputs: subfolders containing the raw .edf files output by the Eyelink 1000+ eye-tracker. .edf files are divided into groups of 25 successive stimuli displayed.</li> <li>Raw_gaze_points: .csv files, containing the raw gaze points locations for each subject, stimulus and mask. Blinks, invalid coordinates and out-of-bounds gaze points are already removed from these files.</li> <li>Fixations: .csv files (one per image per participant) containing eye fixations locations extracted from the raw gaze points using a I-VT algorithm with a saccade velocity threshold of 45 deg/s.</li> </ul> <h2><br><strong>Experiment</strong></h2> <h3>Images</h3> <p>This database consists of 100 images collected from personal collections and various public image datasets, such as the CityScapes and the KITTI-360 datasets, covering a variety of themes, including landscapes, people, actions, and nature.<br>Images had a 1920 x 1080 pixel resolution (FullHD) and were shown on a screen with the same resolution. </p> <h3>Eye-tracking data collection</h3> <p>The stimuli were presented to a group of 37 observers, with normal or corrected-to-normal vision. We used a table-mounted EyeLink 1000 Plus eye-tracker, working at a fixed rate of 1000Hz, with a chin rest to ensure data accuracy. Participants were informed that they would observe images both with and without simulated visual field impairments, and the two types of masks used were described. They were also told that the only task is to view the images freely. </p> <p>To familiarize participants with the different types of masks, four training images (distinct from the trial dataset) were provided in the training phase, with each image presented in the three conditions -- peripheral mask, foveal mask, and control (no mask) -- with varying mask sizes. <br>In the trial phase, stimuli were displayed in a random order, with a randomly generated playlist ensuring no three consecutive stimuli were of the same image. A uniform gray screen was displayed between each stimulus. </p> <p>Viewing distance was set to be 90cm.<br>Eye-tracker calibration was carried out using a 9-point calibration protocol, i.e., 9 points were sequentially and randomly shown on the screen, where the observer should fixate their gaze. <br>Additionally, calibration was performed after every 25 displayed images.</p> <p>Each image was presented to the observers for a 5-second period.</p> <p>Each image in the database was presented to the observers under different conditions :<br> - The image without any obstruction, referred to as the control condition (C). <br> - Peripheral mask (P) simulating a tunnel vision, with two circular mask size variations (one individual mask for each eye): 1.5° and 4.5° radius of field of view, referred to as P1 and P2, respectively. <br> - Foveal mask (F) simulating a central scotoma, also with two size variations: 1.5° and 4.5° radius of obstruction, referred to as F1 and F2, in that order.</p> <p> </p>
Pupil and gaze during binocular rivalry
<p><span><span><span><span><span><span><span><span><span><span><span>The pupil provides a rich, non-invasive measure of the neural bases of perception and cognition, and has been of particular value in uncovering the role of arousal-linked neuromodulation, which alters cortical processing as well as pupil size. But pupil size is subject to a multitude of influences, which complicates unique interpretation. We measured pupils of observers experiencing perceptual multistability -- an ever-changing subjective percept in the face of unchanging but inconclusive sensory input. In separate conditions the endogenously generated perceptual changes were either task-relevant or not, allowing a separation between perception-related and task-related pupil signals. Perceptual changes were marked by a complex pupil response that could be decomposed into two components: a dilation tied to task execution and plausibly indicative of an arousal-linked noradrenaline surge, and an overlapping constriction tied to the perceptual transient and plausibly a marker of altered visual cortical representation. Constriction, but not dilation, amplitude systematically depended on the time interval between perceptual changes, possibly providing an overt index of neural adaptation. These results show that the pupil provides a simultaneous reading on interacting but dissociable neural processes during perceptual multistability, and suggest that arousal-linked neuromodulation shapes action but not perception in these circumstances.</span></span></span></span></span></span></span></span></span></span></span></p>
Developmental Changes in Gaze Behavior and the Effects of Auditory Emotion Word Priming in Emotional Face Categorization
<p>Data used for statistical analyses in the journal article "Developmental Changes in Gaze Behavior and the Effects of Auditory Emotion Word Priming in Emotional Face Categorization" published in the journal Multisensory Research (online publication date: 16 September 2021).</p>
Video demonstration of the Gaze alternation behaviour element in the Interactive group (behaviour element ID code: IIGA) for Korcsok and Korondi (2023), Biologia Futura
<p>The video demonstrates the behaviour element: Gaze alternation in the interactive experimental group, as exhibited by a social robot (behaviour element ID code: <strong>IIGA</strong>). The video is part of an ethogram cataloguing the behaviour elements of the robot, described in the publication: <em><strong>How do you do the things that you do? - Ethological approach to the description of robot behaviour</strong></em> submitted to Biologia Futura (2023) by Korcsok, B. and Korondi, P.</p>
The Effects of a Group-based Gaze Training Intervention for Children With Developmental Coordination Disorder
ClinicalTrials.gov study NCT02904980. IPD Sharing: NO. Countries: 1. Publications: 2.
Evaluating the Validity of an Eye Gaze Paradigm in Predicting Autism Spectrum Disorder
ClinicalTrials.gov study NCT02573428. IPD Sharing: UNDECIDED. Countries: 1. Publications: 19.
Comparison of Exergaming and Vestibular Training on Gaze Stability, Balance and Gait Performance of Older Adults.
ClinicalTrials.gov study NCT04414462. IPD Sharing: NO. Countries: 1. Publications: 6.
Automated Vision Assessment and Impairment Detection Through Gaze Analysis in Wet AMD Patients
ClinicalTrials.gov study NCT06518512. IPD Sharing: NO. Countries: 1. Publications: 10.
RECOGNeyes Gaze-Control Training
ClinicalTrials.gov study NCT06691646. IPD Sharing: NO. Countries: 1. Publications: 2.
Influence of Gaze Shift and Emotions on Symptoms of Blepharospasm
ClinicalTrials.gov study NCT01759745. IPD Sharing: Not stated. Countries: 1. Publications: 5.
A Serious Game to Rehabilitate Gaze Stability in Children with Vestibular Deficit
ClinicalTrials.gov study NCT04353115. IPD Sharing: NO. Countries: 1. Publications: 1.
Effects of Balance Training With Gaze Stabilization Exercises in Elderly Patients With Chronic Dizziness
ClinicalTrials.gov study NCT04751006. IPD Sharing: NO. Countries: 1. Publications: 4.
Oxytocin Modulates Eye Gaze Behavior During Social Processing
ClinicalTrials.gov study NCT03293511. IPD Sharing: Not stated. Countries: 1. Publications: 1.
Effects of Otago and Gaze Stabilization Exercises on Balance, Gait and QOL in Elderly Stroke Patients
ClinicalTrials.gov study NCT06845709. IPD Sharing: NO. Countries: 1. Publications: 3.
Otago Exercise Program And Gaze Stability Exercise In Older Adults
ClinicalTrials.gov study NCT05781776. IPD Sharing: NO. Countries: 1. Publications: 1.
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.