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.
27
datasets available to search
ShareScore release 0.7.1
Dataset results
27 results for “event-based”
PEDRo: an Event-based Dataset for Person Detection in Robotics
<p>This event-based dataset is specifically designed for person detection in robotics applications. </p>
Aircraft Marshaling Signals Dataset of FMCW Radar and Event-Based Camera for Sensor Fusion
<p><strong>Dataset Introduction</strong></p><p>The advent of neural networks capable of learning salient features from variance in the radar data has expanded the breadth of radar applications, often as an alternative sensor or a complementary modality to camera vision. Gesture recognition for command control is arguably the most commonly explored application. Nevertheless, more suitable benchmarking datasets than currently available are needed to assess and compare the merits of the different proposed solutions and explore a broader range of scenarios than simple hand-gesturing a few centimeters away from a radar transmitter/receiver. Most current publicly available radar datasets used in gesture recognition provide limited diversity, do not provide access to raw ADC data, and are not significantly challenging. To address these shortcomings, we created and make available a new dataset that combines FMCW radar and dynamic vision camera of 10 aircraft marshalling signals (whole body) at several distances and angles from the sensors, recorded from 13 people. The two modalities are hardware synchronized using the radar's PRI signal. Moreover, in the supporting publication we propose a sparse encoding of the time domain (ADC) signals that achieve a dramatic data rate reduction (>76%) while retaining the efficacy of the downstream FFT processing (<2% accuracy loss on recognition tasks), and can be used to create an sparse event-based representation of the radar data. In this way the dataset can be used as a two-modality neuromorphic dataset.</p><p><strong>Synchronization of the two modalities</strong></p><p>The PRI pulses from the radar have been hard-wired to the event stream of the DVS sensor, and timestamped using the DVS clock. Based on this signal the DVS event stream has been segmented such that groups of events (time-bins) of the DVS are mapped with individual radar pulses (chirps).</p><p><strong>Data storage</strong></p><p>DVS events (x,y coords and timestamps) are stored in structured arrays, and one such structured array object is associated with the data of a radar transmission (pulse/chirp). A radar transmission is a vector of 512 ADC levels that correspond to sampling points of chirping signal (FMCW radar) that lasts about ~1.3ms. Every 192 radar transmissions are stacked in a matrix called a radar frame (each transmission is a row in that matrix). A data capture (recording) consisting of some thousands of continuous radar transmissions is therefore segmented in a number of radar frames. Finally radar frames and the corresponding DVS structured arrays are stored in separate containers in a custom-made multi-container file format (extension .rad). We provide a (rad file) parser for extracting the data out of these files. There is one file per capture of continuous gesture recording of about 10s.</p><p>Note the number of 192 transmissions per radar frame is an ad-hoc segmentation that suits the purpose of obtaining sufficient signal resolution in a 2D FFT typical in radar signal processing, for the range resolution of the specific radar. It also served the purpose of fast streaming storing of the data during capture. For extracting individual data points for the dataset however, one can pool together (concat) all the radar frames from a single capture file and re-segment them according to liking. The data loader that we provide offers this, with a default of re-segmenting every 769 transmissions (about 1s of gesturing).</p><p><strong>Data captures directory organization (</strong><a href="https://zenodo.org/api/records/10359770/draft/files/radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z/content">radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z</a><strong>)</strong></p><p>The dataset captures (recordings) are organized in a common directory structure which encompasses additional metadata information about the captures.</p><p>dataset_dir/<stage>/<room>/<person>-<gesture>-<distance>/ofxRadar8Ghz_yyyy-mm-dd_HH-MM-SS.rad</p><p>Identifiers</p><ul><li>stage [train, test].</li><li>room: [conference_room, foyer, open_space].</li><li>subject: [0-9]. Note that 0 stands for no person, and 1 for an unlabeled, random person (only present in test).</li><li>gesture: ['none', 'emergency_stop', 'move_ahead', 'move_back_v1', 'move_back_v2', 'slow_down' 'start_engines', 'stop_engines', 'straight_ahead', 'turn_left', 'turn_right'].</li><li>distance: ['xxx', '100', '150', '200', '250', '300', '350', '400', '450'] (in cm). Note that xxx is used for none gestures when there is no person present in front of the radar (i.e. background samples), or when a person is walking in front of the radar with varying distances but performing no gesture.</li></ul><p>The test data captures contain both subjects that appear in the train data as well as previously <i>unseen</i> subjects. Similarly the test data contain captures from the spaces that train data were recorded at, as well as from a new <i>unseen</i> open space.</p><p><strong>Files List</strong></p><p><a href="https://zenodo.org/api/records/10359770/draft/files/radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z/content">radar8Ghz-DVS-marshaling_signals_20220901_publication_anonymized.7z</a></p><p>This is the actual archive bundle with the data captures (recordings).</p><p><a href="https://zenodo.org/api/records/10359770/draft/files/rad_file_parser_2.py/content">rad_file_parser_2.py</a></p><p>Parser for individual .rad files, which contain capture data.</p><p><a href="https://zenodo.org/api/records/10359770/draft/files/loader.py/content">loader.py</a></p><p>A convenience PyTorch Dataset loader (partly Tonic compatible). You practically only need this to quick-start if you don't want to delve too much into code reading. When you init a DvsRadarAircraftMarshallingSignals class object it automatically downloads the dataset archive and the .rad file parser, unpacks the archive, and imports the .rad parser to load the data. One can then <i>request from it </i>a training set, a validation set and a test set as torch.Datasets to work with<i>.</i> </p><p><a href="https://zenodo.org/api/records/10359770/draft/files/aircraft_marshalling_signals_howto.ipynb/content">aircraft_marshalling_signals_howto.ipynb</a></p><p>Jupyter notebook for exemplary basic use of loader.py</p><p><strong>Contact</strong></p><p>For further information or questions try contacting first M. Sifalakis or F. Corradi.</p><p> </p>
Event-Based Velocity Prediction for Spiking Neural Networks
<p>This dataset is intended to be used to predict the velocity based on the event pixels present in the data, given in (t, x, y, p) format alongside a ground truth velocity reading. A novel dataset using people and various objects moving in front of an RGB video camera was created. The positions of each entity and the associated times were captured with a Vicon motion tracking system. These two types of data were calibrated so that the movement in the video matched the measurements recorded by the Vicon system. The types of data collected include two different people carrying a calibrated Vicon Active Wand and moving around the room, a Lambda aerial robot with motion tracking markers that flew around the room, a box with motion tracking markers that were tossed back and forth in the air, and the same box was slid across the floor. The video recording was simulated as event camera data. Each pixel changes state independently of all the other pixels. The Open Event Camera Simulator (ESIM) from the Robotics and Perception Group at the University of Zurich and ETH Zurich is used. The event camera simulator used, ESIM, allows for accurate event simulation data. The ground truth velocities were calculated from the position and timestamps recorded by the Vicon system.</p>
Dissemination of information in event-based surveillance, a case study of Avian Influenza - dataset
<p>This dataset contains a set of tables corresponding to the manual analysis of outbreak-related reports detected by two event-based surveillance tools, PADI-web and HealthMap, supporting the submitted article "Dissemination of information in event-based surveillance, a case study of Avian Influenza".</p> <p>The reports were published between 1 July 2018 and 30<sup>st</sup> June 2019 and described one or several avian influenza outbreaks. We collected 337 reports from PADI-web and 115 from HealthMap. Two epidemiologists identified all the reported events in the news, and classified them as official (notified to the World Organization for Animal Health) or non-official.</p> <p>In order to trace back the source of the event’s information, the epidemiologist manually traced the information pathway of all events mentioned in the PADI-web and HealthMap news. The pathway was deducted from the sources cited in the news. When a source was cited with a hyperlink, we followed the hyperlink to retrace the information pathway as far as possible to the primary source. For each cited source, we created a pair of emitter <em>S<sub>E</sub></em> and receptor sources <em>S<sub>R</sub></em>. We labelled each new source with its type (e.g. online news source, national veterinary authority, etc.). We also recorded their geographical focus (local, national or international) and their specialization in the animal health news coverage (general or specialized).</p> <p>The script for data analyses is available at https://github.com/SarahVal/EBS-network.</p>
An event-based precipitation dataset with life cycle evolution using resilient algorithms
<p>The dataset covers eastern Asia at a temporal range of April to June 2016-2020. We identified initial rain clusters (RCs) from the Global Precipitation Measurement 2ADPR dataset and Mesoscale Convective Systems (MCSs) from the Himawari-8 Advanced Himawari Image gridded product. Based on the contours of the initial RCs and MCSs, we then carried out a series of resilient processes, including filtration, segmentation, and consolidation, to obtain the final RCs. The final RCs had a one-to-one correspondence with the relevant MCS. We extracted the RC area, central location, average radar reflectivity profile, average droplet size distribution profile and other precipitation information from the final RCs and retrieved the life cycle evolution of the MCS area, location, and cloud-top brightness temperature from the corresponding MCSs and tracking algorithms. This dataset facilitates studies of the life cycle evolution of precipitation and provides a good foundation for convection parameterizations in precipitation simulations.</p>
Event-based landslide susceptibility models (Styrian Basin, Austria)
<p><strong>Landslide susceptibility models</strong></p> <p>Landslide susceptibility modes are based on rainfall-triggered landslide events in the Styrian basin, Austria, in 2009 and 2014. Landslide susceptibility models are generalized additive models (GAM). <em>Note: Information on the exact location of landslides has been obscured.</em></p> <p><br> <strong>Uncertainty</strong></p> <p>Posterior simulations of the coefficients using a simple Metropolis Hastings sampler and a Gaussian approximation are available for GAM-Spatial and GAM-SM.</p> <p> </p> <p> </p> <table> <caption><strong>Overview of landslide susceptibility models</strong></caption> <thead> <tr> <th scope="col"><strong>GAM</strong></th> <th scope="col"><strong>Variables</strong></th> </tr> </thead> <tbody> <tr> <td>GAM-Co</td> <td>land surface variables, meteorological variables, geology, LULC</td> </tr> <tr> <td>GAM-SM</td> <td>GAM-Co, soil moisture</td> </tr> <tr> <td>GAM-SM+TC</td> <td>GAM-SM, five-day rainfall > 80 mm top-coded</td> </tr> <tr> <td>GAM-Spatial</td> <td>GAM-SM+TC, Gaussian process smoother</td> </tr> </tbody> </table> <p> </p>
High-throughput event-based and frame-based convolutions for event-cameras
<p>Event cameras are promising sensors for on-line and real-time vision tasks, due to their high temporal resolution, low latency and the elimination of redundant static data. Many vision algorithms use some form of spatial convolution (i.e. spatial pattern detection) as a fundamental component, but additional consideration must be taken for event cameras, as the visual signal is asynchronous and sparse. While elegant methods have been proposed for event-based convolutions, they are unsuitable for real scenarios due to their inefficient processing pipeline, and subsequent low event-throughput. This paper presents an efficient implementation based on decoupling the event-based computations from the computationally heavy convolution ones, increasing the maximum event processing rate by 15.92x, to over 10 million events/second, while still maintaining the event-based paradigm of asynchronous input and output. Results on public datasets with modern 640x480 event-camera recordings show that the proposed implementation achieves real-time processing with minimal impact in the convolution result, while the prior state-of-the-art results in latency of over 1 second per-event.</p>
Event-based hyperspectral EELS: towards nanosecond temporal resolution
<p>Here we present the two data sets presented in the work <a href="https://arxiv.org/abs/2110.01706">Event-based hyperspectral EELS: towards nanosecond temporal resolution</a>. Data was processed using Rust.</p> <p>In ASI Cheetah Timepix3, we have two different kinds of events: electron and TDC events (little-endian). Data chunk package is a 8-byte data that begins with "TPX3". Electron hit is a 8-byte packet that contains "0xb" in 60-63 bits. TDC data packet is a 8-byte packet that contains "0x6" in 60-63 bits. 56-59 bits identify if it is comes from TDC Line 1 or 2 and also identifies if it is a falling or a rising edge.</p>
Asynchronous event-based clustering and tracking for intrusion monitoring in UAS
<p>This dataset describes a collection of rosbag files for event-based intruder monitoring using UAS. A DAVIS346 camera was mounted over a DJI Flamewheel F550 Drone, and an onboard computer recorded the sensor information from the event camera. Each dataset includes events, frames, and IMU measurements. The monitoring scenes were recorded outdoors at the School of Engineering of the University of Seville. In each dataset, an intruder moves and hides from the field of view of the camera simulating a scape-intrusion situation. A total of four monitoring setting were recorded:</p> <p><strong>Daylight monitoring:</strong> A daylight scene for intruder monitoring. An intruder runs and hides behind the objects of the scene to evade the camera field of view.<br> <br> <strong>Night light monitoring:</strong> A monitoring scene during the night without the presence of any artificial light. The low light condition increases the difficulty of monitoring task due to the increment of noisy events.<br> <br> <strong>Multi-target:</strong> An experiment with a suspect and a chaser drone moving in the monitoring area. The drone follows the suspect by simulating a pursuit operation.<br> <br> <strong>Monitoring under illumination changes:</strong> A night scene where the lighting conditions changes by the movement of artificial lights in the scene.</p>
FIGURE 5 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 5. Reconstructions of the biogeographical events of the species group of the E. dandara sp. nov., according to Treefitter. The time of events of each frame is arbitrarily assumed. T1-T7 are in relative time. (A) Reconstruction of p-tree of Fig. 3C (E. cavernicola sister of E. elinae); (B) Reconstruction of p-tree of Fig. 3D (E. aduncus sister of E. cavernicola). White square = duplication, white arrow = switching dispersal, black arrow = dispersal or range expansion. The dark gray shading is the Atlantic Forest, medium gray is Caatinga and light gray is Cerrado savannas.
FIGURE 2 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 2. The area cladograms used in the event analysis, TASS (A), BPA (B) (DaSilva et al., 2015b). PE = Pernambuco; Serg = Sergipe; SEsp = Serra do Espinhaço; BA = Bahia; ES = Espirito Santo; SMSP = Serra do Mar de São Paulo; LSRJ = Litoral Norte de São Paulo; Mnt = Serra da Mantiqueira; Boc = Serra da Bocaina; Org = Serra dos Órgãos; SC= Santa Catarina; PR = Paraná; SSP = Sul de São Paulo; Cha = Chapada Diamantina; Cerr = Cerrado.
FIGURE 7 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 7. Male paratype genitalia (Col. UFPB-OP 454). (A) Dorsal view; (B) Lateral view. Scale 50 µm.
FIGURE 4 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 4. Trackograms showing the significant reconstructions of the biogeographical events for the species group of E. dandara sp. nov., according to Treefitter. Organism phylogeny (p-tree) is represented in bold and the area cladogram (h-tree) in gray. (A) Reconstruction of p-tree of fig. 3C (E. cavernicola sister of E. elinae); (B) reconstruction of p-tree of Fig. 3D (E. aduncus sister of E. cavernicola). Black circle = vicariance, white square = duplication, white arrow = switching dispersal, X = extinction, black arrow = dispersal or range expansion.
FIGURE 3 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 3. (A) Phylogenetic hypotheses of Eusarcus species; the alternative topologies for the clade marked with a single asterisk, which includes the new species, are presented in B, C and D; the alternatives to the polytomy, marked with two asterisks, are discussed in Hara & Pinto-da-Rocha (2010).
FIGURE 6 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 6. Holotype of Eusarcus dandara sp. nov. (MNRJ 7055). (A) Dorsal scutum, habitus; (B) Dorsal scutum, left lateral view; (C) Right coxa and trochanter IV, dorsal view; (D) Right femur IV, dorsal view; (E) Right femur IV, ventral view; (F) Right femur IV, lateral view; (G) Right tibia IV dorsal view; (H) Right tibia IV, ventral view. (I) Right tibia IV, lateral view. Scale 1mm.
FIGURE 1 in Event-based biogeography of Eusarcus dandara sp. nov. (Opiliones: Gonyleptidae), an endemic species of the Northern Atlantic Rainforest of Brazil, and its closely related species
FIGURE 1. Map showing the records of E. dandara sp. nov. and the rest of genus range (shaded distribution). Full line is the Congruence Core and dashed line is the Maximum Region of Endemism of the areas of endemism Pernambuco (PE) and Bahia (BA), according to DaSilva et al. (2015a).
Data underlying "EVE is an open modular data analysis software for event-based localization microscopy"
<p>Data underlying the manuscript "EVE is an open modular data analysis software for event-based localization microscopy"</p> <p>Contains raw EBS-recorded SMLM (eveSMLM) data of DNA-PAINT nanoruler, E.coli cell, and aTubulin network in Cos-7 cells (3D and high density acquisitions).</p>
Event-based vision datasets for pedestrian detection in urban scenarios
<p>Datasets supporting the Spiking Perception and processing for Intelligent Detection of pEdestrians on urban Roads (SPIDER) Project. The codebase is available on Gihub at <a href="https://github.com/th-nuernberg/spider">https://github.com/th-nuernberg/spider</a>.<br><br>In Project SPIDER, we propose a novel solution for urban road surveillance using event-based neuromorphic cameras (i.e. Dynamic Vision Sensors – DVS), neural algorithms, and embedded neuromorphic computing platforms (i.e. Brainchip Akida). The solution is described by rapid detection and identification of abnormal activities, typically describing roadside pedestrian and bicyclist dynamics.<br><br>In order to train the detection system, we decided to use DVS camera events. To also obtain ground truth information we added a traditional CMOS camera on the mount with a preset offset capturing the same field of view. The CMOS frames were only used as support in the labeling process and were not included in the training of the system, the testing, or the evaluation, respectively. We have chosen two locations with different properties of the road, the number of pedestrians and bicyclists, and overall different traffic properties. </p><p><strong>The SPIDER project is an award-winning edge solution for the </strong><a href="https://www.tinyml.org/event/tinyml-hackathon-2023-pedestrian-detection/"><strong>TinyML Vision Zero San Jose Hackathon</strong></a><strong>. The project placed 2nd among 29 teams in the world in the final. The final pitch is available at </strong><a href="https://www.youtube.com/watch?v=ZhBCtfalcOk&t=2872s">https://www.youtube.com/watch?v=ZhBCtfalcOk&t=2872s</a></p><p> </p><p><i>Dataset archive content:</i></p><p> </p><p><strong>Dataset location 1</strong></p><p>• 4 lanes (4 per direction) wide street</p><p>• Location: https://goo.gl/maps/JaYGwaTaBHj5H6SL9</p><p>• 50 kmh (urban) speed limit</p><p>• Near the university campus with Pedestrians (people walking) and bicyclists (people biking, scooting, rolling, etc.)</p><p>• Ideal Operating Environment<br> </p><p><br><strong>Dataset location 2</strong></p><p>• 8 lanes (4 per direction) wide street</p><p>• Location: https://goo.gl/maps/jar6AjysZiM2LP5S7</p><p>• 50 kmh (urban) speed limit</p><p>• Near the main train stations of the city and a location with Pedestrians (people walking, running, or jogging), and cyclists (people biking, scooting, rolling, etc.)<br><br> </p><p><strong>Dataset location 3</strong></p><p>• 6 lanes (3 per direction) wide street on the bridge</p><p>• Location: https://goo.gl/maps/SEEsmpgmLPcD8fG7A</p><p>• 50 kmh (urban) speed limit</p><p>• Near ring street of Munich and a location with Pedestrians (people walking, running, or jogging) and bicyclists (people biking, scooting, rolling, etc.)</p><p>• Night-time data acquisition <br><br> </p>
A worldwide event-based debris-flow barrier dam dataset from 1800 to 2023
<p>Debris-flow barrier dams (DFBDs) not only reshape the original hydrological and geological conditions but may also trigger secondary disasters such as outburst floods and upstream aggradation. Data on DFBDs is crucial for related research. However, there is no dataset for DFBDs has been established. To fill this gap, we adopted a rigorous data collection and validation process, reviewed 2519 literatures and media reports, and constructed a dataset of 555 DFBD events. The dataset comprehensively records 38 attributes of the DFBDs, including location, country, trigger, the date of formation, reliability, debris flow density, debris flow velocity, dam length, height, width, volume, blocking mode, stability, longevity, dam material, lake length, lake volume, lake area, failure mechanism, peak discharge, and loss of life, etc.</p>
Event-Based Computing Packet Storm Dataset - 512 Square Heated Plate
<p>Instrumentation data for a 512x512 heated plate run on the POETS event-based compute platform.</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.