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.
80
datasets available to search
ShareScore release 0.7.1
Dataset results
80 results for “IMU”
MPU9250 MEMS IMU Sine wave acceleration excitation along the Z axis
<p><strong>MPU9250 MEMS IMU Sine wave acceleration excitation along the Z axis</strong></p> <p>The file Met4FOF_mpu9250_Z_Acc_10_hz_250_hz_6rep_ADC.dump contains a dump of the ADC protbuff messages recorded by the Met4FoF dataaqusition unit during the calibration measurement. The ADC is sampled synchronously to the data ready signals of the MPU9250.</p> <p>The file Met4FOF_mpu9250_Z_Acc_10_hz_250_hz_6rep_Sensor.dump contains a dump of the MPU9250 protbuff messages recorded by the Met4FoF dataaqusition unit during the calibration measurement.</p> <p>Met4FOF_mpu9250_Z_Acc_10_hz_250_hz_6rep.xlsx contains the accelerations recorded by the PTB refferenzsystem for each measurement run. The phase is referred to the analog reference values in the channel Data_11 </p> <p>Met4FOF_mpu9250_Z_Acc_10_hz_250_hz_6rep.csv contains the values from the excel table in panda readable form.</p> <p>1FE4_AC_CAL.zip contains various measurements of the ADC transfer function as JSON files.</p>
Multimodal video and IMU kinematic dataset on daily life activities using affordable devices (VIDIMU)
<p>Human activity recognition and clinical biomechanics are challenging problems in physical telerehabilitation medicine. However, most publicly available datasets on human body movements cannot be used to study both problems in an out-of-the-lab movement acquisition setting. The objective of the VIDIMU dataset is to pave the way towards affordable patient tracking solutions for remote daily life activities recognition and kinematic analysis. </p> <p>The VIDIMU dataset includes 54 healthy young adults that were recorded on video and 16 of them were simultaneously recorded using custom IMUs. For each subject, 13 activities were registered using a low-resolution video camera and five Inertial Measurement Units (IMUs). Inertial sensors were placed in the lower or the upper limbs of the subject, respectively for activities that involve movement with the lower or the upper body. Video recordings were postprocessed using the state-of-the-art pose estimator <em>BodyTrack</em> (similar to OpenPose, and included in NVIDIA Maxine-AR-SDK) to provide a sequence of 3D joint positions for each movement. Raw IMU recordings were post-processed to compute joint angles by inverse kinematics with <em>OpenSim</em>. For recordings including simultaneous acquisition of video and IMU data types, these signals were used for data file synchronization. Collected data can be further used in applications related to human activity recognition and biomechanics related experiments in simulated home-like settings.</p> <p> </p> <p> </p>
IMU data captured unobtrusively and in-the-wild by Parkinson's disease patients and healthy controls
<p><strong>DATASET</strong></p> <p>The dataset contains IMU signals captured in-the-wild via the accelerometer sensor embedded in modern smartphones, for the purpose of detecting tremorous episodes, related to Parkinson's Disease (PD). A group of 31 PD patients and 14 Healthy controls contributed accelerometer data using their personal smartphones, for a period spanning many months.Tri-axial acceleration values were recorded automatically whenevera phone call was realized. The recording lasted for 75 seconds at the most. Each phone call thus resulted in one recorded accelerometer signal, also referred to as session. Each subject contributed a different amount of sessions depending on the number of phone calls they realized during the data collection period as well as their participation time (they were free to drop-out at any time). A detailed description of the capturing process as well as analysis results, can be found in the related research article.</p> <p>The data is presented as a list of python dictionaries, stored in a pickle file. Each dictionary in the list, corresponds to one subject and containes the following fields:</p> <p>1. subject_id: scalar<br> A numerical value that uniquely identifies the subject.</p> <p>2. subject_sessions: list of numpy.array<br> A list of numpy arrays of shape (N, 4) that contains the tri-axial accelerometer sessions that the subject contributed. N denotes the total length of the session in samples (which varies from session to session) Column 0 of the array contains the timestamps of the accelerometer samples. Columns 1-3 contain the acceleration values across the x,y,z directions.</p> <p>3. session_datetimes: list of datetime objects <br> A list of datetime objects that denote the capturing date and time of the corresponding entries in the subject_sessions field.</p> <p>4. annotation: dict<br> A dictionary containing the following tremor-related annotation values:<br> * updrs16: scalar int<br> The value related to tremor as described in item 16 of the part II of the MDS-UPDRS scale, as reported by the subject.</p> <p>* updrs20_right: scalar int in range [0, 4]<br> The value related to rest tremor in the right hand as described in item 20 of the part III of the MDS-UPDRS scale, as reported by the attending neurologist.</p> <p>* updrs20_left: scalar int in range [0, 4]<br> Same as above but for left hand.</p> <p>* updrs21_right: scalar int in range [0, 4]<br> The value related to action/postural tremor in the right hand as described in item 21 of the part III of the MDS-UPDRS scale, as reported by the attending neurologist.</p> <p>* updrs21_left: scalar int in range [0, 4]<br> Same as above but for left hand.</p> <p>* sp_expert: scalar int in range [0, 1]<br> A binary tremor annotation created by a group of signal processing experts, upon visually examining the contributed signals in both time and frequency domain and taking into consideration the UDPRS scores of each subject. This was necessary due to the intermittent nature of tremor, as well as a number of considerations related to the in-the-wild nature of the data capturing process. For more details, we refer the reader to the dataset description in the related research article.<br> A '1' value indicates that the subject has tremor.<br> A '0' value indicates that the subject doesn't have tremor.</p> <p>* pd_status: scalar int in range [0, 1]<br> A '1' value indicates that the subject is a PD patient.<br> A '0' value indicates that the subject is a Healthy Control</p> <p>Note: Each annotation value refers to the subject as a whole, and not in any one session.<br> </p> <p><strong>ETHICS & FUNDING</strong></p> <p>The study during which the present dataset was collected is a multi-center study approved in each country available (for more info visit: <a href="http://www.i-prognosis.eu/?page_id=3606">http://www.i-prognosis.eu/?page_id=3606</a>). Informed consent, including permission for third-party access to pseudo-anonymised data, was obtained from all subjects prior to their engagement with the study. The work has received funding from the European Union's Horizon 2020 research and innovation programme under Grant Agreement No 690494 - i-PROGNOSIS: Intelligent Parkinson early detection guiding novel supportive interventions (<a href="http://www.i-prognosis.eu/">i-prognosis.eu</a>).</p> <p> </p> <p><strong>CORRESPONDANCE</strong></p> <p>Any inquiries regarding this dataset should be adressed to:</p> <p>Mr. Alexandros Papadopoulos (Electrical & Computer Engineer, PhD candidate)</p> <p>Multimedia Understanding Groupmug<br> Department of Electrical & Computer Engineering<br> Aristotle University of Thessaloniki<br> University Campus, Building C, 3rd floor<br> Thessaloniki, Greece, GR54124</p> <p>Tel: +30 2310 996359, 996365 <br> Fax: +30 2310 996398<br> E-mail: alpapado@mug.ee.auth.gr</p> <p> </p> <p><strong>LICENSE</strong></p> <p>This is an open access dataset, licensed under Creative Commons Attribution 4.0 International (<a href="https://creativecommons.org/licenses/by/4.0/">https://creativecommons.org/licenses/by/4.0/</a>).</p> <p> </p> <p><strong>WARRANTY</strong></p> <p>This dataset comes without any warranty. Administrators of this dataset can not be held accountable for any damage (physical, financial or otherwise) caused by the use of this dataset. </p>
IMU and marker-based optical motion capture from a humanoid robot
<p>The motion capture contains walking trials from the lower body of the humanoid robot Reem-C from Pal Robotics (Barcelona, Spain). Seven IMUs were attached on the foot, lower leg, upper leg and pelvis segments. IMU data was collected at 100 Hz. Moreover, the robot motion was captured with a marker-based optical system (Qualisys AB, Göteborg, Sweden) at 150 Hz. The focus of the dataset was mainly walking. There are three trials, each with a length of about 6.5 minutes.<br>The dataset contains the definition of the skeleton (segment lengths and coordinate locations), the actual IMU readings and the pose or kinematics from the optical system.</p>
Arm gesture dataset based on IMU data captured from the Technaid human-robot interaction system
<p>Two files with a dataset of ten different/independent hand gestures are provided (seven static gestures and three dynamic gestures). The data were generated in a IMU system with five sensors worn in the right forearm, right arm, chest, left arm and left forearm of a human. The Technaid human-robot interaction system was used to captured the data. The file "datasetStaticGestures.mat" was used to train and test a classifier whose purpose is the recognition of static gestures. On the other hand, the file "datasetDynamicGestures.mat" was used to train and test a classifier whose purpose is the recognition of dynamic gestures. The latter file contains an extra class (gesture) which represents non-gestures.</p>
Brainport, Highway pilot, car in manual mode, IMU detection
<p><strong>Scenario description</strong>:</p> <p>The detection car drives around the track in manual mode, with IMU detection ON.</p> <p><strong>Session description</strong>:</p> <p>25 laps with VW Tiguan on Automotive Campus Test Track</p> <p><strong>Datasets descriptions</strong>:</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_AdasCommand</strong>: Data from the automated driver assistance system</p> <p>Dataset Description This dataset contains the ADAS command in the vehicle</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_Anomaly</strong>: Data sent from detecting vehicle to the service, and from service to vehicles</p> <p>Dataset Description This dataset contains information about all the detected anomalies on the road</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_AnomalyImage</strong>: Data sent from detecting vehicle to service</p> <p>Dataset Description This dataset contains images of the detected anomalies</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_DriverVehicleInteraction</strong>: Data extracted from the CAN of the vehicle</p> <p>Dataset Description This dataset contains e.g. throttlestatus,clutchstatus,brakestatus,brakeforce,wipersstatus,steeringwheel for the vehicle</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_EnvironmentSensorsAbsolute</strong>: Data extracted from the vehicle environment sensors</p> <p>Dataset Description This dataset contains information about detected object, with absolute coordinates</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_EnvironmentSensorsRelative</strong>: Data extracted from the vehicle environment sensors</p> <p>Dataset Description This dataset contains information about detected object, with relative coordinates</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_Hazard</strong>: Data sent from service to vehicle</p> <p>Dataset Description This dataset contains specific information for a vehicle about anomalies and hazards</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_IotVehicleMessage</strong>: Data sent between all devices, vehicles and services</p> <p>Dataset Description Each sensor data submission is a Message. A Message has an Envelope, a Path, and optionally (but likely) Path Events and optionally Path Media. The envelope bears fundamental information about the individual sender (the vehicle) but not to a level that owner of the vehicle can be identified or different messages can be identified that originate from a single vehicle.</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_PositioningSystem</strong>: Data from GPS on the vehicle</p> <p>Dataset Description This dataset contains speed, longitude, latitude, heading from the GPS</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_Vehicle</strong>: Data from the CAN and sensors about the state of the vehicle</p> <p>Dataset Description This dataset contains a.o temperature and battery state of the vehicles</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_VehicleDynamics</strong>: Data from the CAN and sensors about the state of the vehicle</p> <p>Dataset Description This dataset contains a.o accelerations and speedlimit of the vehicle, as observed from the CAN and the external sensors</p>
Brainport, Highway pilot, detection car, manual mode, camera and IMU detection on
<p><strong>Scenario description</strong>:</p> <p>The detection car drives around the track in manual mode, with Camera and IMU detection ON.</p> <p><strong>Session description</strong>:</p> <p>25 laps with VW Tiguan on Automotive Campus Test Track</p> <p><strong>Datasets descriptions</strong>:</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_AdasCommand</strong>: Data from the automated driver assistance system</p> <p>Dataset Description This dataset contains the ADAS command in the vehicle</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_Anomaly</strong>: Data sent from detecting vehicle to the service, and from service to vehicles</p> <p>Dataset Description This dataset contains information about all the detected anomalies on the road</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_AnomalyImage</strong>: Data sent from detecting vehicle to service</p> <p>Dataset Description This dataset contains images of the detected anomalies</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_DriverVehicleInteraction</strong>: Data extracted from the CAN of the vehicle</p> <p>Dataset Description This dataset contains e.g. throttlestatus,clutchstatus,brakestatus,brakeforce,wipersstatus,steeringwheel for the vehicle</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_EnvironmentSensorsAbsolute</strong>: Data extracted from the vehicle environment sensors</p> <p>Dataset Description This dataset contains information about detected object, with absolute coordinates</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_EnvironmentSensorsRelative</strong>: Data extracted from the vehicle environment sensors</p> <p>Dataset Description This dataset contains information about detected object, with relative coordinates</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_Hazard</strong>: Data sent from service to vehicle</p> <p>Dataset Description This dataset contains specific information for a vehicle about anomalies and hazards</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_IotVehicleMessage</strong>: Data sent between all devices, vehicles and services</p> <p>Dataset Description Each sensor data submission is a Message. A Message has an Envelope, a Path, and optionally (but likely) Path Events and optionally Path Media. The envelope bears fundamental information about the individual sender (the vehicle) but not to a level that owner of the vehicle can be identified or different messages can be identified that originate from a single vehicle.</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_PositioningSystem</strong>: Data from GPS on the vehicle</p> <p>Dataset Description This dataset contains speed, longitude, latitude, heading from the GPS</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_Vehicle</strong>: Data from the CAN and sensors about the state of the vehicle</p> <p>Dataset Description This dataset contains a.o temperature and battery state of the vehicles</p> <p><strong>AUTOPILOT_BrainPort_HighwayPilot_VehicleDynamics</strong>: Data from the CAN and sensors about the state of the vehicle</p> <p>Dataset Description This dataset contains a.o accelerations and speedlimit of the vehicle, as observed from the CAN and the external sensors</p>
Wrist-mounted IMU data towards the investigation of free-living human eating behavior - the Free-living Food Intake Cycle (FreeFIC) dataset
<p><strong>Introduction</strong></p> <p>The Free-living Food Intake Cycle (FreeFIC) dataset was created by the <a href="http://mug.ee.auth.gr">Multimedia Understanding Group</a> towards the investigation of <em>in-the-wild</em> eating behavior. This is achieved by recording the subjects’ meals as a small part part of their everyday life, unscripted, activities. The FreeFIC dataset contains the <span class="math-tex">\(3D\)</span> acceleration and orientation velocity signals (<span class="math-tex">\(6\)</span> DoF) from <span class="math-tex">\(22\)</span> in-the-wild sessions provided by <span class="math-tex">\(12\)</span> unique subjects. All sessions were recorded using a commercial smartwatch (<span class="math-tex">\(6\)</span> using the Huawei Watch 2™ and the MobVoi TicWatch™ for the rest) while the participants performed their everyday activities. In addition, FreeFIC also contains the start and end moments of each meal session as reported by the participants.</p> <p><strong>Description</strong></p> <p>FreeFIC includes <span class="math-tex">\(22\)</span> in-the-wild sessions that belong to <span class="math-tex">\(12\)</span> unique subjects. Participants were instructed to wear the smartwatch to the hand of their preference well ahead before any meal and continue to wear it throughout the day until the battery is depleted. In addition, we followed a self-report labeling model, meaning that the ground truth is provided from the participant by documenting the start and end moments of their meals to the best of their abilities as well as the hand they wear the smartwatch on. The total duration of the <span class="math-tex">\(22\)</span> recordings sums up to <span class="math-tex">\(112.71\)</span> hours, with a mean duration of <span class="math-tex">\(5.12\)</span> hours. Additional data statistics can be obtained by executing the provided python script <em>stats_dataset.py</em>. Furthermore, the accompanying python script <em>viz_dataset.py </em>will visualize the IMU signals and ground truth intervals for each of the recordings. Information on how to execute the Python scripts can be found below.</p> <pre><code># The script(s) and the pickle file must be located in the same directory. # Tested with Python 3.6.4 # Requirements: Numpy, Pickle and Matplotlib # Calculate and echo dataset statistics $ python stats_dataset.py # Visualize signals and ground truth $ python viz_dataset.py</code></pre> <p>FreeFIC is also tightly related to Food Intake Cycle (FIC), a dataset we created in order to investigate the <em>in-meal</em> eating behavior. More information about FIC can be found <a href="https://zenodo.org/record/4421861">here</a> and <a href="https://mug.ee.auth.gr/intake-cycle-detection/">here</a>.</p> <p><strong>Publications</strong></p> <p>If you plan to use the FreeFIC dataset or any of the resources found in this page, please cite our work:</p> <pre><code>@article{kyritsis2020data, title={A Data Driven End-to-end Approach for In-the-wild Monitoring of Eating Behavior Using Smartwatches}, author={Kyritsis, Konstantinos and Diou, Christos and Delopoulos, Anastasios}, journal={IEEE Journal of Biomedical and Health Informatics}, year={2020}, publisher={IEEE}}</code></pre> <pre><code>@inproceedings{kyritsis2017automated, title={Detecting Meals In the Wild Using the Inertial Data of a Typical Smartwatch}, author={Kyritsis, Konstantinos and Diou, Christos and Delopoulos, Anastasios}, booktitle={2019 41th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, year={2019}, organization={IEEE}} </code></pre> <p><strong>Technical details</strong></p> <p>We provide the FreeFIC dataset as a <a href="https://docs.python.org/3/library/pickle.html">pickle</a>. The file can be loaded using Python in the following way:</p> <pre><code class="language-python">import pickle as pkl import numpy as np with open('./FreeFIC_FreeFIC-heldout.pkl','rb') as fh: dataset = pkl.load(fh)</code></pre> <p>The dataset variable in the snipet above is a dictionary with <span class="math-tex">\(5\)</span> keys. Namely:</p> <ul> <li>'subject_id'</li> <li>'session_id'</li> <li>'signals_raw'</li> <li>'signals_proc'</li> <li>'meal_gt'</li> </ul> <p>The contents under a specific key can be obtained by:</p> <pre><code class="language-python">sub = dataset['subject_id'] # for the subject id ses = dataset['session_id'] # for the session id raw = dataset['signals_raw'] # for the raw IMU signals proc = dataset['signals_proc'] # for the processed IMU signals gt = dataset['meal_gt'] # for the meal ground truth </code></pre> <p>The <em>sub</em>, <em>ses</em>, <em>raw</em>, <em>proc </em>and <em>gt </em>variables in the snipet above are lists with a length equal to <span class="math-tex">\(22\)</span>. Elements across all lists are aligned; e.g., the <span class="math-tex">\(3\)</span>rd element of the list under the 'session_id' key corresponds to the <span class="math-tex">\(3\)</span>rd element of the list under the 'signals_proc' key.</p> <p><em>sub</em>: list<br> Each element of the sub list is a scalar (integer) that corresponds to the unique identifier of the subject that can take the following values: <span class="math-tex">\([1, 2, 3, 4, 13, 14, 15, 16, 17, 18, 19, 20]\)</span>. It should be emphasized that the subjects with ids <span class="math-tex">\(15, 16, 17, 18, 19\)</span> and <span class="math-tex">\(20\)</span> belong to the held-out part of the FreeFIC dataset (more information can be found in <span class="math-tex">\( \)</span>the publication titled "A Data Driven End-to-end Approach for In-the-wild Monitoring of Eating Behavior Using Smartwatches" by Kyritsis <em>et al).</em> Moreover, the subject identifier in FreeFIC is in-line with the subject identifier in the FIC dataset (more info <a href="https://zenodo.org/record/4421861">here</a> and <a href="https://mug.ee.auth.gr/intake-cycle-detection/">here</a>); i.e., FIC’s subject with id equal to <span class="math-tex">\(2\)</span> is the same person as FreeFIC’s subject with id equal to <span class="math-tex">\(2\)</span>.</p> <p><em>ses: </em>list<br> Each element of this list is a scalar (integer) that corresponds to the unique identifier of the session that can range between <span class="math-tex">\(1\)</span> and <span class="math-tex">\(5\)</span>. It should be noted that not all subjects have the same number of sessions.</p> <p><em>raw</em>: list<br> Each element of this list is dictionary with the 'acc' and 'gyr' keys.<br> The data under the 'acc' key is a <span class="math-tex">\(N_{acc} \times 4\)</span> numpy.ndarray that contains the timestamps in seconds (first column) and the <span class="math-tex">\(3D\)</span> raw accelerometer measurements in <span class="math-tex">\(g\)</span> (second, third and forth columns - representing the <span class="math-tex">\(x, y \)</span> and <span class="math-tex">\(z\)</span> axis, respectively). The data under the 'gyr' key is a <span class="math-tex">\(N_{gyr} \times 4\)</span> numpy.ndarray that contains the timestamps in seconds (first column) and the <span class="math-tex">\(3D\)</span> raw gyroscope measurements in <span class="math-tex">\({degrees}/{second}\)</span>(second, third and forth columns - representing the <span class="math-tex">\(x, y \)</span> and <span class="math-tex">\(z\)</span> axis, respectively). All sensor streams are transformed in such a way that reflects all participants wearing the smartwatch at the same hand with the same orientation, thusly achieving data uniformity. This transformation is in par with the signals in the FIC dataset (more info <a href="https://zenodo.org/record/4421861">here</a> and <a href="https://mug.ee.auth.gr/intake-cycle-detection/">here</a>). Finally, the length of the raw accelerometer and gyroscope numpy.ndarrays is different <span class="math-tex">\((N_{acc} \neq N_{gyr})\)</span>. This behavior is predictable and is caused by the Android platform.</p> <p><em>proc: </em>list<br> Each element of this list is an <span class="math-tex">\(M\times7\)</span> numpy.ndarray that contains the timestamps, <span class="math-tex">\(3D\)</span> accelerometer and gyroscope measurements for each meal. Specifically, the first column contains the timestamps in seconds, the second, third and forth columns contain the <em><span class="math-tex">\(x,y\)</span></em> and <span class="math-tex">\(z\)</span> accelerometer values in <span class="math-tex">\(g\)</span><strong> </strong>and the fifth, sixth and seventh columns contain the <em><span class="math-tex">\(x,y\)</span></em> and <span class="math-tex">\(z\)</span> gyroscope values in <span class="math-tex">\({degrees}/{second}\)</span>. Unlike elements in the <em>raw </em>list, processed measurements (in the <em>proc</em> list) have a constant sampling rate of <span class="math-tex">\(100\)</span> Hz and the accelerometer/gyroscope measurements are aligned with each other. In addition, all sensor streams are transformed in such a way that reflects all participants wearing the smartwatch at the same hand with the same orientation, thusly achieving data uniformity. This transformation is in par with the signals in the FIC dataset (more info <a href="https://zenodo.org/record/4421861">here</a> and <a href="https://mug.ee.auth.gr/intake-cycle-detection/">here</a>). <em>No other preprocessing is performed on the data</em>; e.g., the acceleration component due to the Earth's gravitational field is present at the processed acceleration measurements. The potential researcher can consult the article "A Data Driven End-to-end Approach for In-the-wild Monitoring of Eating Behavior Using Smartwatches" by Kyritsis <em>et al. </em>on how to further preprocess the IMU signals (i.e., smooth and remove the gravitational component).</p> <p><em>meal_gt: </em>list<br> Each element of this list is a<strong> <span class="math-tex">\(K\times2\)</span></strong> matrix. Each row represents the meal intervals for the specific in-the-wild session. The first column contains the timestamps of the meal start moments<strong> </strong>whereas the second one the timestamps of the meal end moments. All timestamps are in seconds. The number of meals <span class="math-tex">\(K\)</span> varies across recordings (e.g., a recording exist where a participant consumed two meals).</p> <p><strong>Ethics and funding</strong></p> <p>Informed consent, including permission for third-party access to anonymised data, was obtained from all subjects prior to their engagement in the study. The work has received funding from the European Union's Horizon 2020 research and innovation programme under Grant Agreement No 727688 - <a href="https://bigoprogram.eu/">BigO: Big data against childhood obesity</a>.</p> <p><strong>Contact</strong></p> <p>Any inquiries regarding the FreeFIC dataset should be addressed to:</p> <p>Dr. Konstantinos KYRITSIS</p> <p>Multimedia Understanding Group (MUG)<br> Department of Electrical & Computer Engineering<br> Aristotle University of Thessaloniki<br> University Campus, Building C, 3rd floor<br> Thessaloniki, Greece, GR54124</p> <p>Tel: +30 2310 996359, 996365 <br> Fax: +30 2310 996398<br> E-mail: kokirits [at] mug [dot] ee [dot] auth [dot] gr</p>
Wrist-mounted IMU data towards the investigation of in-meal human eating behavior - the Food Intake Cycle (FIC) dataset
<p><strong>Introduction</strong></p> <p>The Food Intake Cycle (FIC) dataset was created by the <a href="http://mug.ee.auth.gr">Multimedia Understanding Group</a> towards the investigation of <em>in-meal</em> eating behavior. The FIC dataset contains the triaxial acceleration and orientation velocity signals (<span class="math-tex">\(6\)</span> DoF) from <span class="math-tex">\(21\)</span> meal sessions provided by <span class="math-tex">\(12\)</span> unique subjects. All meals were recorded in the restaurant of Aristotle University of Thessaloniki using a commercial smartwatch, the Microsoft Band <span class="math-tex">\(2\)</span>™ for ten out of the twenty-one meals and the Sony Smartwatch <span class="math-tex">\(2\)</span>™ for the remaining meals. In addition, the start and end moments of each food intake cycle as well as of each micromovement are annotated throughout the FIC dataset.</p> <p><strong>Description</strong></p> <p>A total of <span class="math-tex">\(12\)</span> subjects were recorded while eating their launch at the university’s cafeteria. The total duration of the <span class="math-tex">\(21\)</span> meals sums up to <span class="math-tex">\(246\)</span> minutes, with a mean duration of <span class="math-tex">\(11.7\)</span> minutes. Each participant was free to select the food of their preference, typically consisting of a starter soup, a salad, a main course and a desert. Prior to the recording, the participant was asked to wear the smartwatch to the hand that he typically uses in his everyday life to manipulate the fork and/or the spoon. A GoPro™ Hero <span class="math-tex">\(5\)</span> camera was already set at the table of the participant using a small, <span class="math-tex">\(23\)</span> cm in height, tripod facing the participant, including both the food tray and upper body part in it’s field of view. The purpose of video recording was to obtain ground truth data by manually annotating the IMU sequences based on the video stream. Participants were also asked to perform a clapping hand movement both at the start and end of the meal, for synchronization purposes (as this movement is distinctive in the accelerometer signal). No other instructions were given to the participants. It should be noted that the FIC dataset does not contain instances related with liquid consumption or eating without the fork, knife and spoon (e.g. eating directly with hands). The accompanying python script <em>viz_dataset.py </em>will visualize the IMU signals and food intake cycle (i.e., bite) ground truth intervals for each of the recordings. Information on how to execute the Python scripts can be found below.</p> <pre><code class="language-python"># The script(s) and the pickle file must be located in the same directory. # Tested with Python 3.6.4 # Requirements: Numpy, Pickle and Matplotlib # Visualize signals and ground truth $ python viz_dataset.py</code></pre> <p>FIC is also tightly related to FreeFIC, a dataset we created in order to investigate the <em>in-the-wild </em>eating behavior. More information on FreeFIC can be found <a href="https://zenodo.org/record/4421951">here</a> and <a href="https://mug.ee.auth.gr/free-food-intake-cycle-detection/">here</a>.</p> <p><strong>Annotation</strong></p> <p><em>Micromovements</em></p> <p>For all recordings, the start and end points of all <span class="math-tex">\(6\)</span> micromovements of interest were manually labeled. The micromovements of interest include:</p> <ul> <li><strong>p</strong>ick food, wrist manipulates a fork to pick food from the plate</li> <li><strong>u</strong>pwards, wrist moves upwards, towards the mouth area</li> <li><strong>d</strong>ownwards, wrist moves downwards, away from the mouth area</li> <li><strong>m</strong>outh, wrist inserts food in mouth</li> <li><strong>n</strong>o movement, wrist exhibits no movement</li> <li><strong>o</strong>ther movement, every other wrist movement</li> </ul> <p>The annotation process was performed in such a way that the start and end times of each micro-movement span the whole meal session, without overlapping each other.</p> <p><em>Food intake cycles</em></p> <p>For all recordings, we annotated the start and end points for each intake cycle (i.e. every bite). Each food intake cycle starts with a <strong>p</strong>, ends with a <strong>d </strong>and contains an <strong>m </strong>micromovement.</p> <p><strong>Publications</strong></p> <p>If you plan to use the FIC dataset or any of the resources found in this page, please cite our work:</p> <pre><code>@article{kyritsis2019modeling, title={Modeling Wrist Micromovements to Measure In-Meal Eating Behavior from Inertial Sensor Data}, author={Kyritsis, Konstantinos and Diou, Christos and Delopoulos, Anastasios}, journal={IEEE journal of biomedical and health informatics}, year={2019}, publisher={IEEE}}</code></pre> <pre><code>@inproceedings{kyritsis2017food, title={Food intake detection from inertial sensors using lstm networks}, author={Kyritsis, Konstantinos and Diou, Christos and Delopoulos, Anastasios}, booktitle={International Conference on Image Analysis and Processing}, pages={411--418}, year={2017}, organization={Springer}}</code></pre> <pre><code>@inproceedings{kyritsis2017automated, title={Automated analysis of in meal eating behavior using a commercial wristband IMU sensor}, author={Kyritsis, Konstantinos and Tatli, Christina Lefkothea and Diou, Christos and Delopoulos, Anastasios}, booktitle={2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, pages={2843--2846}, year={2017}, organization={IEEE}}</code></pre> <p><strong>Technical details</strong></p> <p>We provide the FIC dataset as a <a href="https://docs.python.org/3/library/pickle.html">pickle</a>. The file can be loaded using Python in the following way:</p> <pre><code class="language-python">import pickle as pkl import numpy as np with open('./FIC.pkl','rb') as fh: dataset = pkl.load(fh)</code></pre> <p>The <em>dataset </em>variable in the snipet above is a dictionary with <span class="math-tex">\(6\)</span> keys. Namely:</p> <ul> <li>'subject_id'</li> <li>'session_id'</li> <li>'signals_raw'</li> <li>'signals_proc'</li> <li>'meal_gt'</li> <li>'bite_gt'</li> </ul> <p>The contents under a specific key can be obtained by:</p> <pre><code>sub = dataset['subject_id'] # for the subject id ses = dataset['session_id'] # for the session id raw = dataset['signals_raw'] # for the raw IMU signals proc = dataset['signals_proc'] # for the processed IMU signals mm = dataset['mm_gt'] # for the micromovement ground truth bite = dataset['bite_gt'] # for the bite ground truth</code></pre> <p>The <em>sub</em>, <em>ses</em>, <em>raw</em>, <em>proc, mm </em>and<em> gt </em>variables in the snipet above are lists with a length equal to <span class="math-tex">\(21\)</span>. Elements across all lists are aligned; e.g., the <span class="math-tex">3</span>rd element of the list under the 'session_id' key corresponds to the <span class="math-tex">3</span>rd element of the list under the 'signals_proc' key.</p> <p><em>sub</em>: list<br> Each element of the sub list is a scalar (integer) that corresponds to the unique identifier of the subject that can take values between <span class="math-tex">\(1\)</span> and <span class="math-tex">\(12\)</span>. Moreover, the subject identifier in FIC is in-line with the subject identifier in the FreeFIC dataset (information available <a href="https://zenodo.org/record/4421951">here</a> and <a href="https://mug.ee.auth.gr/free-food-intake-cycle-detection/">here</a>); i.e., FIC’s subject with id equal to <span class="math-tex">2</span> is the same person as FreeFIC’s subject with id equal to <span class="math-tex">2</span>.</p> <p><em>ses: </em>list<br> Each element of this list is a scalar (integer) that corresponds to the unique identifier of the session that can range between <span class="math-tex">1</span> and <span class="math-tex">\(3\)</span>. It should be noted that not all subjects have the same number of sessions.</p> <p><em>raw</em>: list<br> Each element of this list is dictionary with the 'acc', 'gyr' and 'offset' keys.<br> The data under the 'acc' key is a <span class="math-tex"><span class="math-tex">\(N_{acc}\times4\)</span></span> numpy.ndarray that contains the timestamps in seconds (first column) and the <span class="math-tex"><em><span class="math-tex">\(3D\)</span></em></span> raw accelerometer measurements in <span class="math-tex">\(g\)</span> (second, third and forth columns - representing the <span class="math-tex">\(x, y\)</span> and <span class="math-tex"><em><span class="math-tex">\(z\)</span></em></span> axis, respectively). The data under the 'gyr' key is a <span class="math-tex"><span class="math-tex">\(N_{gyr} \times 4\)</span></span> numpy.ndarray that contains the timestamps in seconds (first column) and the <span class="math-tex">\(3D\)</span> raw gyroscope measurements in <span class="math-tex">\(degrees/second\)</span>(second, third and forth columns - representing the<span class="math-tex"><em> <span class="math-tex">\(x, y\)</span></em></span> and <span class="math-tex">\(z\)</span> axis, respectively). All sensor streams are transformed in such a way that reflects all participants wearing the smartwatch at the same hand with the same orientation, thusly achieving data uniformity. This transformation is in par with the signals in the FreeFIC dataset (information available <a href="https://zenodo.org/record/4420039">here</a> and <a href="https://mug.ee.auth.gr/free-food-intake-cycle-detection/">here</a>). Finally, the length of the raw accelerometer and gyroscope numpy.ndarrays is different <span class="math-tex"><span class="math-tex">\(N_{acc} \neq N_{gyr}\)</span></span>. This behavior is predictable and is caused by the Android/MS Band platforms. The offset key contains a float that is used to align the IMU sensor streams with the videos that were used for annotation purposes (videos are not provided).</p> <p><em>proc: </em>list<br> Each element of this list is an <span class="math-tex">\(M \times 7\)</span> numpy.ndarray that contains the timestamps, <span class="math-tex"><em><span class="math-tex">\(3D\)</span></em></span> accelerometer and gyroscope measurements for each meal. Specifically, the first column contains the timestamps in seconds, the second, third and forth columns contain the <span class="math-tex">\(x,y\)</span> and <span class="math-tex"><em><span class="math-tex">\(z\)</span></em></span> accelerometer values in <span class="math-tex"><em><span class="math-tex">\(g\)</span></em></span><strong> </strong>and the fifth, sixth and seventh columns contain the <span class="math-tex">\(x, y\)</span> and <span class="math-tex"><em><span class="math-tex">\(z\)</span></em></span> gyroscope values in <span class="math-tex"><em><span class="math-tex">\(degrees/second\)</span></em></span>. Unlike elements in the <em>raw </em>list, processed measurements (in the <em>proc</em> list) have a constant sampling rate of <span class="math-tex">100</span> Hz and the accelerometer/gyroscope measurements are aligned with each other. In addition, all sensor streams are transformed in such a way that reflects all participants wearing the smartwatch at the same hand with the same orientation, thusly achieving data uniformity. This transformation is in par with the signals in the FreeFIC dataset (information available <a href="https://zenodo.org/record/4420039">here</a> and <a href="https://mug.ee.auth.gr/free-food-intake-cycle-detection/">here</a>). <em>No other preprocessing is performed on the data</em>; e.g., the acceleration component due to the Earth's gravitational field is present at the processed acceleration measurements. The potential researcher can consult the article "Modeling Wrist Micromovements to Measure In-Meal Eating Behavior from Inertial Sensor Data" by Kyritsis <em>et al. </em>on how to further preprocess the IMU signals (i.e., smooth and remove the gravitational component).</p> <p><em>mm</em>: list<br> Each element of this list is a <span class="math-tex">\(K \times 3\)</span> numpy.ndarray. Each row represents a single micromovement interval. The first column contains the timestamps of the start moments in seconds, the second column the timestamps of the end moments in seconds and the third column a number representing the type of the micromovement. The identifier to micromovement mapping is provided below:<br> <span class="math-tex">\([1] \rightarrow\)</span> <strong>n</strong>o movement<br> <span class="math-tex">\([2] \rightarrow\)</span> <strong>u</strong>pwards<br> <span class="math-tex">\([3] \rightarrow\)</span> <strong>d</strong>ownwards<br> <span class="math-tex">\([4] \rightarrow\)</span> <strong>p</strong>ick food<br> <span class="math-tex">\([5] \rightarrow\)</span> <strong>m</strong>outh<br> <span class="math-tex">\([6] \rightarrow\)</span> <strong>o</strong>ther movement</p> <p><em>bite</em>: list<br> Each element of this list is a <strong><span class="math-tex">\(L\times2\)</span></strong> numpy.ndarray. Each row represents a single food intake event (i.e., a bite). The first column contains the start moments while the second column contains the end moments of each intake event. Both the start and end moments are provided in seconds.</p> <p><strong>Ethics and funding</strong></p> <p>Informed consent, including permission for third-party access to anonymised data, was obtained from all subjects prior to their engagement in the study. The work has received funding from the European Union's Horizon 2020 research and innovation programme under Grant Agreement No 727688 - <a href="https://bigoprogram.eu/">BigO: Big data against childhood obesity</a>.</p> <p><strong>Contact</strong></p> <p>Any inquiries regarding the FIC dataset should be addressed to:</p> <p>Dr. Konstantinos KYRITSIS</p> <p>Multimedia Understanding Group (MUG)<br> Department of Electrical & Computer Engineering<br> Aristotle University of Thessaloniki<br> University Campus, Building C, 3rd floor<br> Thessaloniki, Greece, GR54124</p> <p>Tel: +30 2310 996359, 996365 <br> Fax: +30 2310 996398<br> E-mail: kokirits [at] mug [dot] ee [dot] auth [dot] gr</p>
Camera-IMU calibration with rolling shutter camera
<p>This video illustrates the performance of the proposed rolling shutter camera-IMU calibration. A checkerboard is used to help the monocular camera recover the scale so that the relative translation can also be calibrated.</p> <p>The red pluses show the detected checkerboard corners, while the green squares show the projection using the filtered camera pose.</p> <p>As can be seen, in the case of motion blur, the detected corners may not be accurate with some drifts, while the projected corners are still close to the real corners.</p>
In-vehicle Sensing Datasets (e.g., GPS, IMU, and OBD data) In Florida
<p>This data collection and distribution is supported by NSF OAC-1948066. These datasets include a total of 497 trajectory datasets over 2404 km. Each dataset includes 6DOF IMU data (e.g., triaxial acceleration and gyroscope data), GPS data (e.g., latitude, longitude, altitude, speed over ground, the number of connected satellites, Course Over Ground), and OBD data (e.g., rpm, throttle positions, accelerator positions, RPM, air temperature, etc.). The data collection mechanism adopts the asynchronous sampling technologies that make capturing sensor data independent of the recorded signal. Therefore, datasets collected from each sensor are logged in separate files (e.g., time_obd.jsonl, time_gps.jsonl, time_obd.jsonl). By matching the time when each sensor module initiated to log data, one can aggregate/fuse multi-type in-vehicle sensing data.</p><p> </p>
Wrist and Tibia/Shoe Mounted IMU Measurement Results for Gait Analysis
<p>This document describes a dataset of measurement results collected using wrist, tibia and shoe mounted inertial sensors. The main purpose of the dataset was to test signal translation algorithms converting signals registered using the wrist-worn sensor e.g. a smartwatch to signals which would be measured with a shoe or tibia - mounted device. The dataset includes tri-axial acceleration and angular velocity registered during several walks.</p> <p>More details are included in the dataset_description pdf file.</p> <p>When using the data, please consider also citing the original paper, for which it was collected:</p> <p>Kolakowski, M.; Djaja-Josko, V.; Kolakowski, J.; Cichocki, J. Wrist-to-Tibia/Shoe Inertial Measurement Results Translation Using Neural Networks. <em>Sensors</em> <strong>2024</strong>, <em>24</em>, 293. <a href="https://doi.org/10.3390/s24010293" target="_blank" rel="noopener">https://doi.org/10.3390/s24010293</a></p> <p><strong>The dataset will be gradually updated as new data are gathered.</strong></p>
Dataset for Monitoring and Visualizing Stroke Rehabilitation Progress using Wearable Sensors (IMU)
<div> <p>This dataset is associated with a manuscript that is currently under peer review.</p> <p> </p> <p>Article Abstract:</p> <p>Stroke is one of the leading causes of death and disability worldwide, and recovering mobility is an important goal during post-stroke rehabilitation. In this work, we present a study to verify the feasibility of monitoring and visualizing longitudinal stroke gait rehabilitation progress using wearable sensors. Wearable devices such as inertial measurement units (IMUs) are easy-to-use and cost-effective tools for quantifying mobility. However, there is a need for research on longitudinal monitoring of stroke rehabilitation progress with wearables, as well as generating clinically relevant insights using appropriate visualizations. To this aim, we recruited ten stroke patients in their early rehabilitation stage. We collected and analyzed the IMU-derived gait features across two visits, and presented visualizations of the foot movement trajectories as well as the spatio-temporal gait parameters in the average, symmetry, and variation domains to quantify changes in gait. Our visualization and quantification methods are evaluated and validated by clinical experts, and prove to be promising in aiding clinicians to monitor rehabilitation progression.</p> <p> </p> <p>Data description:</p> <p>The dataset consists data from ten stroke patients who completed both visits. The "raw" data folder contains tri-axial acceleration and angular velocity data from the IMUs. In addition, information about the participants such as demographics (e.g., body height and body weight), FAC scores at both visits, and evaluations of gait improvement are documented in the file "participant_info.csv".</p> <p>The “interim” folder contains IMU data that has been manually segmented to remove irrelevant movements before and after each walking session during a visit, based on visual inspection of raw IMU signals. For quality control, the segmented accelerometer and gyroscope data of each sensor were plotted, and the plots were saved in the same folder as the IMU signals. In addition, during the first execution of gait parameter extraction, calculated 3D feet trajectories were cached in the "interim" folder, so that for future executions, the cached trajectories can be loaded directly, reducing the computational efforts for re-calculation. The file "stance_magnitude_thresholds_manual.csv" documents the angular velocity thresholds used to identify stance phases for the gait analysis algorithm for each participant. The threshold values were determined manually by observing the angular velocity signals. </p> <p>The “processed” folder contains stride-by-stride spatio-temporal gait parameters extracted for each of the four walking conditions, and aggregated gait parameters in terms of coefficients of variation and symmetry for all walking conditions for each participant. </p> <p> </p> </div>
IMU-Based Tip-Over Dataset for Space Exploration Rover Dynamics and Stability Analysis
<p>This dataset provides a comprehensive collection of Inertial Measurement Unit (IMU) sensor data captured from a space exploration rover under varying conditions of terrain, speed, and inclination. The primary goal of this dataset is to enable the study of dynamic stability, specifically the detection and analysis of tip-over events, which are critical for safe and efficient operation of autonomous rovers in extraterrestrial environments.</p> <p> </p> <p><em>This dataset is provided by the Robotics Innivation Center, DFKI GmbH.</em></p> <p><em>The grant was provided by Federal Ministry for Economic Affairs and Climate Action </em></p> <p><em>Grant number: 50RA2124</em></p>
IMU data from walking trials on a treadmill
<p>This dataset contains inertial measurement unit data from 10 treadmill walking trials. The subject is of legal age and wore a total of 12 Xsens MTw Awinda IMUs. More details in our publication (https://doi.org/10.7717/peerj.15097).</p> <p>Includes a readme file with details.</p>
Indoor localization using Wi-Fi and IMU at National Taiwan University CSIE 5F
<p>A dataset composed of Wi-Fi fingerprints and IMU sensing data which collect by smartphone.</p> <p>We collected this dataset at National Taiwan University CSIE building 5F.</p> <p>The txt files are raw data.</p> <p>Fingerprint.txt is the Wi-Fi fingerprint set for reference map.</p> <p>Track1.txt and Track2.txt are the Wi-Fi fingerprints and IMU sensing data collect by android smartphone.</p> <p> </p> <p>The .npy files are the preprocessed data.</p> <p> </p> <p> </p> <p> </p>
Vibration and IMU Sensing Human Activity Dataset
<p>This dataset contains fine-grained human daily activity data collected by infrastructure vibration sensors and one on-wrist IMU sensor. This dataset is collected from six persons from two domestic homes, in total, there are 12 sub-datasets.</p> <p>For the naming, "p" means person and "l" means location.</p> <p>Each dataset has 11 columns, 1o of them stands for sensors' reading.</p> <p>* Due to the uploading platform, please<strong> <em>ignore</em> </strong>all files in the folder '__MACOSX', and files whose names start with '._'. These are computer system files, not parts of the shared dataset. </p> <p>** If you are going to use this dataset for any publications, we will appreciate you to cite this dataset properly.</p> <p>************************************************************</p> <p>The following content is copied from README.txt in the compressed folder:</p> <p>-----------------------<br> Labels:</p> <p>Keyboard typing 1<br> Using mouse 2<br> Handwriting 3<br> Cutting vegetables 4<br> Stir-frying vegetables 5<br> Wiping the table 6<br> Sweeping floor 7<br> Using vacuum to vacuum floor: 8<br> Open and close drawer: 9</p> <p>None Activity: 10</p> <p>-----------------------<br> 11 Columns:<br> 1: Activity label<br> 2: Vibration sensor put on the Living Area floor<br> 3: Vibration sensor put on the Living Area table<br> 4: Vibration sensor put on the Studying Area floor<br> 5: Vibration sensor put on the Studying Area desk<br> 6, 7, 8: Accelerometer X,Y,Z<br> 9, 10, 11: Gyroscope X,Y,Z</p> <p>-----------------------<br> All signals are zero-meaned.<br> The vibration sensors' sampling rate is roughly around 6500Hz, and the IMU sensors' original sampling rate is roughly around 235Hz.</p> <p>************************************************************</p> <p>New in Version 2:</p> <p>- Added extracted features from IMU data and vibration data for reference.</p> <p>- IMU signal is applied with a sliding window of 1.5 seconds with 0.75 seconds overlapping, then the feature is extracted in each window. The feature's description can be found here: https://dl.acm.org/doi/abs/10.1145/3410530.3414320</p> <p>- The vibration signal is applied with event detection to extract events in the vibration signal. For each event, we normalize it by its energy, then extract 10~490 Hz frequency amplitude as the feature.</p> <p> </p> <p>Disclaimer: Both event detection and feature extraction are empirical, we don't guarantee it is an optimal one.</p>
Comparison IMU vs Optotrak
<p>This is a dataset of IMU and optical motion capture data recorded on 16 healthy subjects while performing a series of upper body movements and exercises. A detailed explanation of the setup and experimental protocol is provided within the folder. In short, the goal of this experiment was to assess the accuracy and precision of a series of IMU calibration protocols against a gold standard optical system.</p>
Datasets and Supporting Materials for the IPIN 2018 Competition Track 4 (Foot-Mounted IMU based Positioning, off-site)
<p>This package contains the datasets and supplementary materials used in the IPIN 2018 Competition (Nantes, France).</p> <p><strong>Contents:</strong></p> <ol> <li>IPIN2018_CallForCompetition_v2.1: Call for competition including the technical annex describing the competition </li> <li>01-Logfiles: This folder contains 2 zip files.<br> - HKB08.zip : for sensors bias estimation.<br> - HKB82.zip : for trajectory estimation.<br> Each archive contains 4 files :<br> - HKBxx_mag.csv : magnetometer data<br> - HKBxx_sti.csv : inertial data<br> - HKBxx_ublox.ubx : GNSS data<br> - HKBxx_INFO.txt : info file<br> see page 16 of IPIN2018_CallForCompetition_v2.1.pdf for more details.</li> <li>02-Supplementary_Materials: This folder contains the datasheet files of the different sensors.</li> <li>03-Evaluation: This folder contains the scripts used to calculate the competition metric, the 75th percentile on all evaluation points. The ground truth is provided csv file.</li> </ol> <p><strong>Please, cite the following works when using the datasets included in this package:</strong></p> <ul> <li>Ortiz, M.; Perul, J.; Torres-Sospedra, J. Renaudin, V. Datasets and Supporting Materials for the IPIN 2018 Competition Track 4 (Foot-Mounted IMU based Positioning, off-site), Zenodo 2018 <a href="http://dx.doi.org/10.5281/zenodo.3228012">http://dx.doi.org/10.5281/zenodo.3228012</a></li> </ul> <p><strong>Additional information can be found at:</strong></p> <ul> <li><a href="http://evaal.aaloa.org/2018/call-for-competitions">http://evaal.aaloa.org/2018/call-for-competitions</a></li> <li><a href="http://ipin-conference.org/2018/ipincompetition/">http://ipin-conference.org/2018/ipincompetition/</a></li> </ul> <p><strong>For any further questions about the database and this competition track, please contact to: </strong></p> <ul> <li> <p>Miguel Ortiz (<a href="mailto:miguel.ortiz@ifsttar.fr">miguel.ortiz@ifsttar.fr</a>) at the French Institute of Science and Technology for Transport, Development and Networks (IFSTTAR) France.</p> <p> </p> </li> </ul>
IMU, magnetometer, and motion capture data from a UAV used for indoor magnetic field mapping and localization
<p>IMU, magnetometer, and motion capture data from a UAV used for indoor magnetic field mapping and localization. </p> <p>This data is split up into two folders. </p> <p>"stationary_magnetometer_data" 0.1Hz samples of a RM3100 magnetometer that was kept in one position from July 2022 through February 2023. The data is partitioned into separate files for analytical convenience of our research. Each file here is a CSV with a timestamp (synchronized with chrony to a central computer) and the three components of the measured magnetic field. We did not calibrate this stationary magnetometer.</p> <p>"UAV_and_mocap_data" has many subfolders. Each subfolder is labeled by a date and a small description of the goals for that test segment. There is a single "EXPLANATION" file in each subfolder that gives more detail on the provided data. The data here includes the trajectory flown by the UAV, outdoor calibration data to adjust the raw magnetometer measurements, and IMU/motion capture data for each listed flight test. The EXPLANATION file should explain what trajectory was flown for each individual flight test. </p>
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.