Research Directory

iniVation Cameras (DVS, DAVIS)

Pioneer sensors from Institute of Neuroinformatics. DVS128, DAVIS240/346 series.

Prophesee Metavision Sensors

High-resolution event cameras with advanced software ecosystem. Up to 1280×720 resolution.

Samsung & Sony Development

Next-generation prototypes: Samsung VGA DVS, Sony 35.6Mpixel hybrid sensors, OmniVision 3-wafer stacked designs.

CelePixel CeleX Series

High-resolution sensors including CeleX-V 1 Megapixel event camera.

Intel Loihi

Neuromorphic research chip for spiking neural networks and event-based processing.

DYNAP (aiCTX AG)

Dynamic Neuromorphic Asynchronous Processor with 256 neurons and 128K synapses.

IBM TrueNorth & SpiNNaker

Large-scale neuromorphic computing platforms for event-driven applications.

DDD20 - End-to-End Event Camera Driving Dataset

Fusing frames and events for improved steering prediction

DSEC - Stereo Event Camera Dataset

Large-scale stereo dataset for driving scenarios

Prophesee Automotive Detection Dataset

Large scale event-based detection dataset for automotive applications

N-MNIST & N-Caltech101

Neuromorphic versions of classic vision datasets

N-CARS Dataset

Large real-world event-based dataset for car classification

DVS128 Gesture Dataset

Gesture recognition dataset used in IBM's neuromorphic system

MVSEC - Multi Vehicle Stereo Event Camera

Event camera dataset for 3D perception

VECtor - Versatile Event-Centric Benchmark

Multi-sensor SLAM dataset with IMU, LiDAR, and cameras

M3ED - Multi-Robot Multi-Sensor Multi-Environment

Large-scale multi-modal dataset for robotics

jAER (Java Address-Event Representation)

Real-time sensory-motor processing for event-based sensors

Prophesee OpenEB

Open source library for event-based vision

Tonic

Event datasets and transforms library (like Torchvision for events)

RPG DVS ROS

ROS driver and tools for dynamic vision sensors

Prophesee ROS Wrapper

ROS driver and messages for Prophesee event-based sensors

EV-FlowNet

Self-supervised optical flow estimation for event cameras

E2VID

Video reconstruction from event cameras

Event Feature Tracking

Probabilistic data association for feature tracking

Picking Up Quantization Steps for Compressed Image Classification

Li Ma, Peixi Peng, Guangyao Chen, Yifan Zhao, Siwei Dong, Yonghong Tian

Novel approach to reduce neural network sensitivity to compressed images by utilizing quantization steps from compressed files. Proposes quantization aware confidence (QAC) and quantization aware batch normalization (QABN) for improved classification performance.

PID-NET: A Novel Parallel Image-Dehazing Network

Wei Liu, Yi Zhou, Dehua Zhang, Yi Qin

Advanced image dehazing network combining CNN and Vision Transformer architectures with lightweight attention mechanisms and redundant feature filtering for superior performance in various atmospheric conditions.

DQnet: Cross-Model Detail Querying for Camouflaged Object Detection

Raphael Berner et al.

Cross-model detail querying approach for detecting camouflaged objects in complex visual environments, advancing computer vision capabilities for challenging detection scenarios.

Deep Learning Methods for Fruit Fly Regurgitation Studies

Tongzhou Zhou, Wei Zhan, Mengyuan Xiong

Comprehensive application of deep learning and computer vision techniques for studying fruit fly regurgitation behavior, including I3D behavior recognition, U-Net segmentation with CBAM attention, and YOLOv5+DeepSort trajectory tracking.

Q-YOLO: Efficient Inference for Real-time Object Detection

Qing Niao et al.

Optimized YOLO architecture for efficient real-time object detection with improved inference speed and accuracy, focusing on practical deployment scenarios.

Emerging Topics in Computer Vision and AI

Recent advances by our research team span multiple domains including neuromorphic vision systems, event-based processing, biological behavior analysis, and agricultural AI applications.

Interdisciplinary Research Impact

Our collaborations bridge computer science with agricultural sciences, biological research, and industrial applications, demonstrating the versatility of modern AI techniques across diverse scientific domains.

Institute of Neuroinformatics (INI)

University of Zurich & ETH Zurich | Pioneer in neuromorphic vision sensors

Robotics and Perception Group (RPG)

University of Zurich | Event-based vision for high-speed robotics

Robotic Interactive Perception

TU Berlin | Event-based robot vision algorithms

Event-Driven Perception for Robotics (EDPR)

Istituto Italiano di Tecnologia (IIT) | Bio-inspired robotics

GRASP Lab

University of Pennsylvania | Kostas Daniilidis's group

Perception and Robotics Group

University of Maryland | Fermüller's Lab on event-based vision

Intel Labs

Neuromorphic Computing | Mike Davies - Loihi development

Nano(neuro)electronics Research Lab

Purdue University | Neuromorphic hardware research

International Centre for Neuromorphic Systems

Western Sydney University | Australia

Camera Intelligence Lab

Peking University | China

Mobile Perception Lab

ShanghaiTech University | China

Visual Intelligence Lab

KAIST | South Korea

Event-based Robot Vision (TU Berlin)

Comprehensive course with YouTube videos and slides

Event Camera Tutorial (Tobi Delbruck)

Introductory video series about DVS technology

Telluride Neuromorphic Workshops

Annual workshops on neuromorphic engineering

Event-based Algorithms for Geometric Computer Vision

Alex Zhu, University of Pennsylvania, 2019

Event Cameras: from SLAM to High Speed Video

Henri Rebecq, University of Zurich, 2019

How to See with an Event Camera

Cedric Scheerlinck, Australian National University, 2021