Home

Awesome

<h1 align="center"> Event-based Vision / Dynamic Vision Sensors / Neuromorphic Vision / Event Camera - Resources </h1>

This repository is dedicated to categorizing articles on event-based vision. While many papers are directly linked to PDFs hosted on platforms like arXiv or OpenAccess, certain articles may necessitate access through an academic license. These include publications from IEEE, Springer, Elsevier journals, and others. (Last Update on 9/19/2024)

Search Keywords:

Event-based vision, event-based camera, DVS, dynamic vision sensor, neuromorphic sensor, event camera

Github Page Curators:

<a href="https://chakravarthi589.github.io/" target="_blank"> Bharatesh Chakravarthi, Ph.D<br> <a href="https://aayush-v.github.io/" target="_blank"> Aayush Atul Verma

Our Publications:

<p align="center"> <img src="cvpr24_eTram.png" alt="Event Cameras" width="400px"> <img src="cvpr24_SEVD.png" alt="Event Cameras" width="400px"> </p>

<a id="event-based-cameras"></a>

:high_brightness: Event-based Cameras

<p align="center"> <img src="eventCameras.png" alt="Event Cameras" width="500px"> </p>
CompanyCameras / DevicesSoftwares/Tools
<a href="https://inivation.com/" target="_blank"> iniVation </a>DAVIS346 , DAVIS346 AER, DVXplorer, DVXplorer Micro <br/> DVXplorer Lite DVXplorer S Duo, Stereo Kit <br/> <a href="https://inivation.com/wp-content/uploads/2023/11/2023-11-iniVation-devices-Specifications.pdf" target="_blank"> Camera Specifications / Comparisions </a><a href="https://docs.inivation.com/software/introduction.html" target="_blank"> Software Solutions </a>
<a href="https://www.prophesee.ai/" target="_blank"> Prophesee </a>Metavision® EVK4 – HD, Metavision® EVK3 – GENX320, <br/> Metavision® EVK3 – HD, Metavision Starter Kit–AMD Kria KV260, <br/> Metavision Starter Kit –STM32F7<a href="https://docs.prophesee.ai/stable/index.html" target="_blank"> Metavision SDK </a>
<a href="https://thinklucid.com/triton2-evs/" target="_blank"> Lucid Vision Labs - Triton2 EVS </a>Triton2 EVS 0.9 MP Model (IMX636), Triton2 EVS 0.3 MP Model (IMX637)

:high_brightness: Tools / Simulators

Event-Camera Simulators
<b> DAVIS Simulator: </b> An Event Camera Simulator based on Blender <a href="https://arxiv.org/pdf/1610.08336" target="_blank">[Paper]</a> <a href="https://github.com/uzh-rpg/rpg_davis_simulator" target="_blank">[Github]</a> <a href="https://rpg.ifi.uzh.ch/davis_data.html" target="_blank">[Webpage]
<b> ESIM: </b> An Open Event Camera Simulator <a href="https://rpg.ifi.uzh.ch/docs/CORL18_Rebecq.pdf" target="_blank">[Paper]</a> <a href="https://github.com/uzh-rpg/rpg_esim" target="_blank">[Github]</a> <a href="https://rpg.ifi.uzh.ch/esim.html" target="_blank">[Webpage]
<b> v2e: </b> From Video Frames to Realistic DVS Events <a href="https://tub-rip.github.io/eventvision2021/papers/2021CVPRW_V2E_From_Video_Frames_to_Realistic_DVS_Events.pdf" target="_blank">[Paper]</a> <a href="https://github.com/SensorsINI/v2e" target="_blank">[Github]</a> <a href="https://sites.google.com/view/video2events/home" target="_blank">[Webpage]
<b> ICNS Simulator: </b> Improved Event Camera Simulator via Characterized Parameters <a href="https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2021.702765/full" target="_blank">[Paper]</a> <a href="https://github.com/neuromorphicsystems/IEBCS" target="_blank">[Github]</a> 
<b> V2CE Toolbox: </b> Video to Continuous Events Simulator The <a href="https://arxiv.org/pdf/2309.08891" target="_blank">[Paper]</a> <a href="https://github.com/ucsd-hdsi-dvs/V2CE-Toolbox" target="_blank">[Github]</a> 
<b> DVS-Voltmeter: </b> Stochastic Process-based Event Simulator for Dynamic Vision Sensors <a href="https://www.ecva.net/papers/eccv_2022/papers_ECCV/papers/136670571.pdf" target="_blank">[Paper]</a> <a href="https://github.com/Lynn0306/DVS-Voltmeter" target="_blank">[Github]</a>
<b> Carla: </b> DVS Camera <a href="https://carla.readthedocs.io/en/latest/ref_sensors/#dvs-camera" target="_blank">[Web Link]</a>
<b> Prophesee: </b> Video to Event Simulator <a href="https://docs.prophesee.ai/stable/samples/modules/core_ml/viz_video_to_event_simulator.html" target="_blank">[Web Link]</a>

:high_brightness: CV Conference Papers

CVPRWACVECCVICCV
CVPR 2024 <br> CVPR 2023 <br>CVPR 2022 <br>CVPR 2021 <br>CVPR 2020 <br>CVPR 2019 <br>CVPR 2018 <br>CVPR 2019 <br>CVPR 2017 <br> CVPR 2016WACV 2024 <br> WACV 2023 <br>WACV 2022 <br>WACV 2021 <br>WACV 2020 <br>WACV 2019 <br>WACV 2018 <br>WACV 2019 <br>WACV 2017ECCV 2024 <br> ECCV 2022 <br>ECCV 2020 <br>ECCV 2018 <br> ECCV 2016ICCV 2023 <br>ICCV 2021 <br>ICCV 2019 <br>ICCV 2017

:high_brightness: Event-based Vision Workshop Papers


:high_brightness: Datasets

<!-- - Our Datasets <br/> | Our Datsets | Dataset Description | Type | Event-Modality | Annotations | Object Class | Lighting | Weather | |-------|------|----------|---------------|-------------|-----------------|------------------|--------------------| | <a href="https://eventbasedvision.github.io/eTraM/" target="_blank"> eTraM </a> | Fixed perception traffic monitoring data (10 Hrs) | Real-world | Prophesee EVK4 HD | 2D Bounding Box (~2M BB) | Vehicle, Pedestrian, Micro-mobility | Daytime, Nighttime, Twilight | Clear, Rainy | | <a href="https://eventbasedvision.github.io/SEVD/" target="_blank"> SEVD | Fixed (27 Hrs) and Ego (31 Hrs) perception Multiview Multi-Modality (Event, RGB, Depth, Optical, Semantic, Instance) traffic data | Synthetic | Carla DVS | 2D & 3D Bounding Box (~9M BB) | Vehicle, Pedestrian, Micromo-bility | Daytime, Nighttime, Twilight, Domain shifts | Clear, Cloudy, Wet, Soft Rainy, Hard Rainy, Foggy | -->

:high_brightness: Other Papers

Papers from Google Scholar using the keywords related to event-based vision


:high_brightness: Other Resources


:high_brightness: Survey/Review Papers

CVPR 2024

CVPR 2023

CVPR 2022

CVPR 2021

CVPR 2020

CVPR 2019

CVPR 2018

CVPR 2017

CVPR 2016


WACV 2024

WACV 2023

WACV 2022

WACV 2021

WACV 2020

WACV 2019

WACV 2018

WACV 2017


ECCV 2024

ECCV 2022

ECCV 2020

ECCV 2018

ECCV 2016


ICCV 2023

ICCV 2021

ICCV 2019

ICCV 2017


ECCV 2024 Wokshop on Neuromorphic Vision

ECCV 2024 Wokshop on Neural Fields Beyond Cams


CVPRW 2023

CVPRW 2021

CVPRW 2019


Realworld Event-based Datasets

YearDatasetSensorAnnotation TypeObjectsTaskDescription
2024<a href="https://eventbasedvision.github.io/eTraM/" target="_blank"> eTraM </a>Prophesee EVK4–HDYes, Manual 2D bounding boxVehicles, Pedestrians, Micro-mobilityObject DetectionHigh-resolution, large-scale, fixed traffic perception dataset for traffic monitoring
2024<a href="https://github.com/Event-AHU/EventVOT_Benchmark?tab=readme-ov-file#dvd-eventvot-dataset" target="_blank"> EventVOT </a>Prophesee EVK4–HDYes, Manual annotationsUAV’s, Pedestrians, Vehicles, Ball sportsObject TrackingLarge-scale, high-resolution, visual object-tracking dataset
2024<a href="https://vlislab22.github.io/ExACT/" target="_blank"> SeAct </a>DAVIS346YesHuman actions like sit, catch, throw, vomit, handshakeAction RecognitionEvent-text action recognition dataset
2024<a href="https://www.aimodels.fyi/papers/arxiv/led-large-scale-real-world-paired-dataset" target="_blank"> LED </a>Prophesee EVK4–HDYesAerial targets, pedestrians, vehiclesDenoising / ClassificationPaired real-world event-denoising dataset
2023<a href="https://github.com/SSIGPRO/PEDRo-Event-Based-Dataset" target="_blank"> PEDRo </a>DAVIS346Yes, Manual labeling – 2D bounding boxPersonsObjected detection (person)Large person detection dataset recorded with a moving camera
2022<a href="https://sites.google.com/view/event-based-lipreading" target="_blank"> DVS-Lip </a>DAVIS346YesDifferent sequences containing all words in English vocabularyGesture Recognition (Lip-reading)Lip-reading dataset with over 100 classes, 40 speakers, 19K plus utterances in the English language
2022<a href="https://better-flow.github.io/evimo/download_evimo_2.html" target="_blank"> EVIMO2 </a>Prophesee Gen3, DVS Gen3Yes, Automatic labelingMoving objectsMotion Segmentation, Motion Recognition, Structure from MotionIndoor dataset with moving objects, 3D motion of the sensor, 3D motion of all objects, structure of the object and scene, masks of the objects, and optical flow
2021<a href="https://dsec.ifi.uzh.ch/" target="_blank"> DSEC </a>Prophesee Gen 3.1 sensorYes, Lidar measurements as ground truthDriving scenarios with diverse illumination during day and nighttimeStereo MatchingHigh-resolution stereo dataset for driving scenarios
2020<a href="https://www.prophesee.ai/2020/01/24/prophesee-gen1-automotive-detection-dataset/" target="_blank"> ATIS Automotive Detection (GEN1) </a>Prophesee GEN1 sensorYes, Manual 2D bounding boxCars, PedestriansObject detection, tracking, optical flowLarge event-based automotive (cars and pedestrians) detection dataset
2020<a href="https://www.prophesee.ai/2020/11/24/automotive-megapixel-event-based-dataset/" target="_blank"> 1 Mpx </a>Prophesee 1 megapixel event cameraYes, Automated labeling – 2D bounding boxPedestrians, two-wheelers, cars, trucks, buses, traffic signs, traffic lightsObject detectionHigh-resolution large-scale automotive detection dataset (25M BB, 14 Hours)
2019<a href="https://github.com/PIX2NVS/NVS2Graph" target="_blank"> ASL-DVS </a>DAVIS240cYes24 classes correspond to 24 letters (A-Y, excluding J)Object ClassificationDataset for American Sign Language (ASL) recognition
2019<a href="https://better-flow.github.io/evimo/" target="_blank"> EVIMO </a>DAVIS346Yes, Automatic labelingMoving objectsMotion segmentation, structure from motion, visual odometry, optical flow, stereoIndoor dataset with moving objects, 3D motion of the sensor, 3D motion of all objects, object structure, and object masks
2018<a href="https://www.prophesee.ai/2018/03/13/dataset-n-cars/" target="_blank"> N-Cars </a>Prophesee GEN1 sensorYes, Semi-automatic labeling – 2D bounding boxCarsObject ClassificationLarge real-world event-based dataset for object classification
2018<a href="https://daniilidis-group.github.io/mvsec/" target="_blank"> MVSEC </a>DAVIS346BYesPoses and depthFeature tracking, visual odometry, stereo depth estimationSynchronized stereo pair large dataset recorded from a handheld rig, hexacopter, top of a car, and on a motorcycle, in different illumination levels and environments
2017<a href="https://docs.google.com/document/d/1HM0CSmjO8nOpUeTvmPjopcBcVCk7KXvLUuiZFS6TWSg/pub" target="_blank"> DDDD17 </a>DAVIS346BYesVehicle speed, driver steering etc.,Steering wheel angle predictionAnnotated real-world driving recordings in highways, city during day, evening, night, dry, and wet weather conditions
2017<a href="https://research.ibm.com/publications/a-low-power-fully-event-based-gesture-recognition-system" target="_blank"> DvsGesture </a>DVS128Yes11 Hand and arm gesturesGesture Recognition (Hand and arm)Gesture dataset comprising 11 hand gesture categories from 29 subjects under 3 illumination conditions
2017<a href="https://rpg.ifi.uzh.ch/davis_data.html" target="_blank"> Event Camera Dataset </a>DAVIS 240CYesObjects rotation, translation, person walking and running, etc.,Pose estimation, visual odometry, SLAMThe dataset consists of object's motion captured in outdoor and indoor scenarios with varying speed and DoFs

Synthetic Event-based Datasets

YearDatasetEvent Modality UsedSubject/Object ClassesLabelingTaskDescription
2024<a href="https://eventbasedvision.github.io/SEVD/" target="_blank"> SEVD </a>CARLA DVSCar, truck, van, bicycle, motorcycle, pedestrianYes, Automatic labeling, 2D and 3D BBObject detection, trackingMulti-view ego and fixed perception dataset for traffic monitoring
2024Event-KITTIV2EVehicles, pedestrians, cyclists, etc.,YesObject Detection, trackingAn event-based version of KITTI using the V2E for daytime images and a noise model for nighttime images of corresponding daytime
2023<a href="https://rpg.ifi.uzh.ch/esfp.html" target="_blank"> ESfP-SyntheticESIMScenes consisting of a textured mesh illuminated with a point light sourceYes, ground-truth surface normal from rendererObject reconstructionDataset for Surface normal estimation using event-based shape from polarization
2022N-EPIC-KitchenESIM8 action classes (Put, take, open, close, wash, cut, mix, pour)YesAction RecognitionEvent-based camera extension of the large-scale EPIC-Kitchens dataset
2021<a href="https://github.com/82magnolia/n_imagenet" target="_blank"> N-ImageNetSamsung DVS Gen31000 Object Classes (same as ImageNet)YesObject RecognitionEvent-based version of the original ImageNet dataset.
2017<a href="https://figshare.com/s/d03a91081824536f12a8" target="_blank"> CIFAR10-DVSDVS camera10 Object Classes (airplane, automobile, bird, cat, deer, dog, frog, horse, ship, truck)YesObject ClassificationEvent-based representations of the original CIFAR-10 images
2015<a href="https://www.garrickorchard.com/datasets/n-mnist" target="_blank"> N-MNISTDVS camera10 classes of digits (0-9)YesObject ClassificationEvent-based version of the original MNIST dataset
2015<a href="https://www.garrickorchard.com/datasets/n-caltech101" target="_blank"> N-Caltech 101ATIS image sensor101 Object Classes (animals, vehicles, etc.,)YesObject ClassificationEvent-based version of the traditional Caltech101 dataset
2015<a href="http://www2.imse-cnm.csic.es/caviar/MNISTDVS.html" target="_blank"> MNIST-DVSDVS camera10 classes of digits (0-9)YesObject ClassificationEvent-based version of the original MNIST dataset

Other Papers

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

<!-- [⤴](#event-based-cameras) -->