Davide Scaramuzza
@davsca1
Followers
14,815
Following
163
Media
307
Statuses
626
Professor of Robotics, University of Zurich
Zurich
Joined July 2013
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Telegram
• 1337018 Tweets
#FREEDUROV
• 294996 Tweets
Hezbollah
• 144063 Tweets
#ดาราล้อกันเล่นxชาล็อต
• 98093 Tweets
Lebanon
• 86228 Tweets
嫌いボタン
• 67615 Tweets
#MarkTuanxVALORANT
• 42229 Tweets
Mark Tuan Superpower Stage
• 26670 Tweets
FortPeat FanParty
• 26135 Tweets
新潟2歳
• 25389 Tweets
GULF DHEVAPROM FANCON
• 20894 Tweets
アルコス
• 15876 Tweets
サトノレーヴ
• 15165 Tweets
ビシエド
• 14418 Tweets
#SFormula
• 12863 Tweets
キーンランドカップ
• 11510 Tweets
Pinned Tweet
We are thrilled to share our groundbreaking paper published today in
@Nature
: "Champion-Level Drone Racing using Deep Reinforcement Learning." We introduce "Swift," the first autonomous vision-based drone that beat human world champions in several fair head-to-head races! PDF
42
343
1K
@DisneyResearch
introduces their new robot at
#IROS2023
! Trained in simulation with
#reinforcementlearning
!
@ieeeiros
106
2K
6K
We release our newest work on event cameras: "Time Lens". We use events to upsample low-framerate RGB HD video by over 50 times with only 1/40th of the memory footprint!
#CVPR2021
Paper, code, datasets:
@DanielGehrig6
@MathiasGehrig
12
196
893
Check out our
#CVPR2023
paper Recurrent Vision Transformers for Object Detection with
#eventcameras
! We achieve sota performance (47.2% mAP) while reducing inference time by 6x (<12ms) & improving parameter efficiency 5x! Paper, Code, Video:
@MathiasGehrig
5
118
625
We are releasing the source code of
#UltimateSLAM
, which combines
#events
, frames, and IMU to achieve robust visual
#SLAM
in high speed and high dynamic range scenarios! Paper & Code:
Video:
W/
@RosinolToni
@supitalp
@DanielGehrig6
5
131
535
How can we combine the advantages of both
#ReinforcementLearning
and Model Predictive Control?
Check our
#TRO
paper on “Policy Search for Model Predictive Control with Application to Agile Drone Flight”!
Paper, Code, Full Video: Kudos to
@realyunlong
6
76
512
I am happy to share paper, dataset, code, video "Dronet: Learning to Fly by Driving", where we use Deep Neural Networks to teach drones to navigate autonomously in the streets of a city by imitating cars and bicycles, thus respecting traffic rules!
9
221
490
Check out our
#ICRA2024
paper "Actor-Critic Model Predictive Control." Model-free
#reinforcementlearning
(RL) is known for its strong task performance and flexibility in optimizing general reward formulations. On the other hand,
#ModelPredictiveControl
(MPC) benefits from
4
79
449
Why can
#ReinforcementLearning
(RL) achieve results beyond
#OptimalControl
(OC) in many real-world robotics control tasks? We investigate this question in our paper published today in
@SciRobotics
, available Open Access: . Video:
7
82
443
We are releasing our full quadrotor control framework as open-source. This framework has been used for the past 6 years in my lab. It has served as the basis in over 500 public live demonstrations over 50 scientific publications from my lab:
4
124
392
Can an inexpensive, off-the-shelf IMU be the only sensor to estimate the full state (position, velocity, orientation) of a quadrotor flying through a track at high speed and even be on-pair with vision-based localization? The answer is yes, within certain limitations! In this
6
55
322
Robotics is difficult!
10
35
264
We are thrilled to share our breakthrough research on "Agile Flight from Pixels without State Estimation," to be presented and live-demonstrated at
#RSS2024
next week! You heard well: no state estimation means no explicit visual localization, no SLAM, no VIO, and no IMU!
Paper:
5
47
252
We are excited to release open-source SVO Pro, the latest version of SVO developed over the past 7 years in my lab! Includes sliding window-based backend, global bundle adjustment with loop closure, different camera models, active exposure control!
1
57
247
Honored to receive a Best Paper Award from the IEEE Robotics and Automation Letters for our paper showing how to use
#eventcameras
to prevent
#drones
from crashing when subject to rotor failure! Paper&Code:
@giov_cioffi
@sihao_sun
@UZH_en
@ieee_ras_icra
6
30
245
Check our new paper "Redesigning SLAM for Arbitrary Multi-Camera Systems": novel generic visual SLAM front-end for arbitrary multicamera systems. No need to tune the parameters across different camera setups!
PDF:
@ManasiMuglikar
1
54
242
Check out our
#RAL
#IROS
paper "Learning Minimum-Time Flight in Cluttered Environments". We combine classical path planning and deep
#reinforcementlearning
to train collision-free flight at unprecedented speeds!
Paper:
Video:
1
36
233
We are excited to announce our latest paper in Science Robotics on agile flight allowing us to generate "time-optimal quadrotor trajectories" that are faster than human
#FPV
#droneracing
pilots! Paper: Full video:
Kudos
@foehnph
5
30
233
Upon many requests, I am releasing the PPT file (with all videos embedded) of my 2-hour tutorial on event cameras (I updated it today). The file is 2.7GB, has 113 slides, 40 videos, and links to all external material. I hope you find this useful.
6
48
221
Nature did amazing work putting up this 5-minute video explanation of how our Swift AI drone works and how it combines data from simulation and the physical world to train the AI controller, and why simulation alone is not enough. It features an interview with drone-racing world
5
38
221
If you are at
#RSS2024
come and see our live demo on how to fly fast from pixels without state estimation (no visual localization, no SLAM, no VIO, no IMU). Last chance today from 12:00 to 13:30 hrs in the Mechanical Engineering building of TU Delft.
Paper:
6
29
215
We release E-RAFT: Dense Optical Flow from Event Cameras! 3x lower error, robust to low-light and HDR! We also release an Optical Flow Benchmark with automatic code evaluation and leader board! Paper, Code, Data:
@MathiasGehrig
@marayjay
@DanielGehrig6
0
43
210
We are hiring multiple PhD students and Postdocs:
Are you ready to take your research career to the next level? We are seeking highly motivated PhD students and postdoctoral researchers to join our lab and make a significant impact on real-world
1
44
214
We are releasing the code of our Fisher Information Field, the first dedicated map for perception-aware planning that is >10x faster than standard methods using point clouds!
Paper:
Code:
Video:
@ZichaoZhang6
2
47
203
Want to track features in high-speed motion? Check out our
#CVPR2023
paper and Award candidate, which introduces the first data-driven feature tracker for
#eventcameras
Paper, code:
, Kudos
@Messikommer1
@carter_fang
@MathiasGehrig
4
46
196
We are truly honored to receive the
#IROS2023
Best Paper Award for our paper “Autonomous Power Line Inspection with Drones via Perception-Aware MPC”! Congratulations to my coauthors Jiaxu Xing
@jixing24
, Giovanni Cioffi
@giov_cioffi
, and Javier Hidalgo-Carriò!
Paper:
6
11
200
We are excited to announce our survey paper on Autonomous Drone Racing, published in the IEEE Transactions on Robotics (T-RO)! The survey covers the latest developments in agile flight for both model-based and learning-based approaches, from modeling to perception, state
2
40
200
We are excited to announce our new, indoor, drone-testing arena! Equipped with 36
@Vicon
cameras and with 30x30x8m of flight space, we can deploy advanced perception, learning, control algorithms to push vision-based agile drones to speeds over 60 km/h and accelerations over 5g!
7
36
197
Computing time-optimal trajectories for
#quadcopters
can now be done in just 10ms and onboard! This helps withstand both wind and dynamic environments! Check out our TRO paper and its follow-up:
#UZH_en
#drones
#droneracing
#control
3
36
192
An end-to-end policy trained in simulation flies vision-based
#drones
*in the wild* at up to 40 kph! Check out our new
@SciRobotics
paper. Code released! w/
@antoniloq
@kaufmann_elia
Video:
Paper, Code & Datasets:
@UZH_en
#Intel
6
40
189
Are you looking for a photo-realistic, customizable, and easy to use simulator for
#drones
? Then,
#Flightmare
is what you need! Compatible with
#ROS
,
#OpenAI
Gym, and
#Oculus
.
Website:
Kudos to
@Cprimer_
@kaufmann_elia
@antoniloq
#UZH_en
1
51
179
Stanford University, Microsoft Research, and myself are organizing an Autonomous Drone Racing Competition at NeurIPS! APply here:
#UZH_eng
#droneracing
#deeplearning
#AI
#NeurIPS2019
4
53
172
We are launching a new robot competition to push the limits on planning, perception, and control in highly dynamic environments. The winner will get a
@SkydioHQ
2 drone and an invited talk at our
#ICRA2021
workshop w/
@antoniloq
@lucacarlone1
@MarkusRyll
2
35
171
Congratulations to my student Yunlong Song
@realyunlong
for successfully defending his Ph.D. in “Learning Robot Control: From
#ReinforcementLearning
to
#DifferentiableSimulation
”! Many thanks to the external reviewers Yuke Zhu
@yukez
from the U. of Texas at Austin, Marco Hutter
8
11
163
Can you guess who wrote a paper just by reading it? We present a
#transformer
-based AI that achieves over 70% accuracy on the newly created, largest-to-date, authorship-attribution dataset with over 2000 authors!
Kudos to
@l_bauersfeld
@ManasiMuglikar
4
26
160
This past weekend, we celebrated the 25th anniversary of Roland Siegwart’s Autonomous Systems Lab. A lab and a man who have shaped the world of robotics. Over a hundred former lab members joined this memorable event. Thanks
@rsiegwart
and happy
@ASL_ETHZ
anniversary to all!
2
8
161
We have PhD and Postdoc openings at the ETH AI Center. If you wish to work with me, please apply **simultaneously** through the center and also through my lab:
Deadlines: for PhD positions: Nov. 30, 2021, for Postdocs, Dec. 15, 2021.
Only 2 weeks left to apply for the ETH AI Center Doctoral Fellowships 2022!
Ready to shape the future of
#AI
and
#MachineLearning
?
Apply now to the ETH AI Center Doctoral or Postdoctoral Fellowship 2022! More information on:
#phd
#postdoc
#education
#ML
1
14
27
3
33
151
Drones with on-board sensing and computation can now fly agile acrobatic maneuvers!
Check out our
#RSS2020
paper "Deep Drone Acrobatics"!
Paper/Code:
Video:
Kudos:
@antoniloq
@kaufmann_elia
@UZH_en
#Intel
#MachineLearning
#drones
4
49
151
We release open source our event-camera simulator, called ESIM: ESIM will allow you to unleash the potential of event cameras without the need to buy one!
#learning
#artificialintelligence
#deeplearning
#drones
#robot
#camera
#UZH
#DVS
#SLAM
#eventcamera
0
58
151
Honored to receive the
#RSS2024
Best Demo Paper Award for our paper “Demonstrating Agile Flight from Pxels without State Estimation”! Big congrats to the team
@isgeles
@l_bauersfeld
@roaguiangel
@jixing24
@realyunlong
Maria Krinner!
Paper:
Narrated video:
4
17
151
We are releasing the source code of EVO, our monocular Event-based Visual Odometry described in the RAL'17 paper! We hope this code will help novice researchers to enter the exciting field of
#eventcamera
#SLAM
! Code, data, paper:
@DanielGehrig6
@supitalp
0
24
148
Check out our RAL/
#ICRA2021
paper "Human-Piloted Drone Racing: Visual Processing and Control”. We show how eye movements relate to flight behavior in human drone racing pilots! We release the dataset!
Paper:
Talk:
@chr_pfeiffer
3
35
148
Check out our paper on event-driven visual control of drones on the
#Intel
#Loihi
#neuromorphic
chip, where a
#Spiking
#NeuralNetwork
directly maps
#DVS
events to control commands! Done with
@Intel
!
Paper:
Video:
w/
@sandamirskaya
0
30
146
Autonomous systems are revolutionizing industries from surveying to delivery. But to navigate as agile as humans, we need robust algorithms. Learn about the progression of this field and the challenges ahead in our new survey on
#autonomousdroneracing
3
37
144
Check out our
#PAMI
paper with code "Dense Continuous-Time Optical Flow from Event Cameras," where we show how to regress *continuous-time* trajectories of every pixel from event cameras alone or events plus frames! The key idea is to iteratively estimate per-pixel polynomials
1
14
145
We are open-sourcing a toolbox to facilitate event camera calibration! Neural network-based image reconstruction enables compatibility with any existing calibration toolbox!
Code:
Paper:
@ManasiMuglikar
@MathiasGehrig
@DanielGehrig6
2
33
144
Check out our
#ICRA2024
paper "Contrastive Initial State Buffer for Reinforcement Learning," which tackles the sample inefficiency in
#ReinforcementLearning
head-on. Code released! We introduce an approach agnostic to the underlying RL algorithm: the Contrastive Initial State
1
30
143
I am excited to announce our latest work on
#agile
#quadrotor
control under uncertainty. L1-NMPC allows us to fly high-speed trajectories with unknown payloads, including ice-cold beer!
Paper:
Video:
W/
@foehnph
@kaufmann_elia
3
24
140
I am proud to announce that my team Philipp Föhn Davide Falanga Julien Kohler Elia Kaufmann won the IROS2018 Autonomous Drone Race Competition, completing the track in just 30 sec!
#IROS2018
#drones
#DeepLearning
3
30
136
If you are in Zurich this Thursday, November 2, at 17:30hrs, feel free to join my talk at
@ETH_en
about our most recent research on agile flight, from perception to action, from frames to event cameras. The talk will take place in room ML D28. I will cover the recently published
2
9
132
We are excited to announce the
#SwissDroneDays
: June, 11-12, 2022, Zurich! We will attempt a world record with an autonomous AI-powered
#drone
. Info:
#robotics
#machinelearning
#deeplearning
#computervision
0
27
125
The journal of Field Robotics is back and is the first fully open access journal in Robotics!
Announcing Field Robotics, a new Open Access journal dealing with the fundamentals of robotics in unstructured and dynamic environments. Papers are now being accepted.
1
63
253
1
16
129
We are releasing the UZH-FPV Drone Racing Dataset: Contain event cameras, standard cameras, IMU, and ground truth from FPV drone flown by professional pilots! VIO competition with money prize coming up, stay tuned!
#droneracing
#VIO
#visualodometry
2
32
128
Check out our RAL paper "Autonomous Quadrotor Flight despite Rotor Failure with Onboard Vision Sensors: Frames vs. Events"! Control and VIO code released!
Paper:
Code:
Video:
Kudos to
@sihao_sun
@giov_cioffi
3
35
126
Congrats to my student Henri Rebecq
@supitalp
for successfully defending his PhD on Event Cameras: from SLAM to High Speed Video! Thanks to the external reviewers
@AjdDavison
, T. Delbruck, B. Schiele! Here is a video recap of Henri's memorable contributions!
#UZH_en
#eventcameras
2
23
127
DSEC, our new event-based car dataset is out! >400GB of data: stereo VGA
#Prophesee
cameras, stereo RGB cameras,
#Velodyne
lidar, HDR, day and night, urban and mountain driving. Check it out and participate in our
#CVPR2021
competition:
Kudo
@MathiasGehrig
3
23
126
Excited to announce and release open-source EDS: Event-aided Direct Sparse Odometry!
#CVPR2022
(oral). First direct
#visualodometry
method combining
#events
and frames! Paper, Code, Datasets:
Kudos
@jhidalgocarrio
@guillermogb
@UZH_en
#eventcameras
#slam
2
31
123
So honored that our
#CVPR2021
on video frame interpolation with event cameras is on
#twominutepapers
!
@DanielGehrig6
@StepanTulyakov
2
17
126
We are excited to share our
#CVPR2024
paper "State Space Models for Event Cameras"! Code released! Traditional deep neural networks struggle with event-camera data at higher frequencies, leading to a trade-off between accuracy and speed. We introduce
#StateSpaceModels
(
#SSMs
) to
2
18
123
Check out our PAMI paper on
#sim2real
transfer for High Speed and High Dynamic Range Video Reconstruction with an
#EventCamera
! We show we can render high speed phenomena, e.g. bullet hitting an object!
Paper, code and datasets:
@supitalp
#UZH_en
3
21
118
We are thrilled to announce a Special Issue on Visual SLAM in the IEEE Transactions on Robotics (T-RO). Submission deadline: December 15, 2024. We invite contributions on latest advancements on topics related but not limited to:
- Robust SLAM
- Representations in SLAM (NeRF,
1
24
114
NASA just released the video of the landing, which is btw based on computer vision via Terrain Relative Navigation: onboard images are matched against those from the Mars orbiter both for VIO and for scouting landing spots: more info
0
21
115
Check out our
#3DV2021
paper on Event-based Structure Light! We combine an
#eventcamera
with a laser-point projector to allow accurate depth estimation in highly dynamic scenes! Paper, Code, Data: Kudos to
@ManasiMuglikar
@guillermogb
@Prophesee_ai
3
20
115
Can we learn a 3D perception system with the same inputs as an human, i.e. vision and sparse depth measurements? Yes! To know how, see our paper by
@antoniloq
&
@Google
"Learning Depth with Very Sparse Supervision"
PDF:
Video:
1
16
113
I was honored to receive the IEEE Kiyo Tomiyasu Technical Field Award last week during the
#ICRA2024
conference “for contributions to agile visual navigation of micro drones and low-latency robust perception with event cameras”! This is the third time this prestigious IEEE career
11
10
116
Today we did our first experiment in reduced, hyper, and zero gravity! Our goal: study how different g affects self motion estimation in drone pilots in view of future human space missions. A unique opportunity made possible by the
@UZHspacehub
@UZH_en
@NLR_NL
w/
@chr_pfeiffer
1
6
112
Autonomous cars equipped with event cameras are now possible with our CARLA simulator plugin!
Available since CARLA v0.9.10
@carlasimulator
.
Code:
Doc:
Kudos:
@jhidalgocarrio
#carlasimulator
#autonomousdriving
#eventcameras
#UZH
1
24
114
Check out our work with
@NASAJPL
on the first Data-Efficient Collaborative Decentralized Thermal-Inertial Odometry
#IROS2022
! Paper, Code, Video, Datasets: Kudo
@polivicio
,
@neorobo
,
@jhidalgocarrio
@jeffdelaune
2
20
108
Do you want to push the boundary of aerial robotics? Participate in our
#icra2022
#DodgeDrone
Challenge! .
We provide an easy-to-use API and a
#ReinforcementLearning
framework! The winner will get a talk invitation and a money prize!
1
23
111
Check out our
@cvpr2019
paper "Events to Video: Bringing Modern Computer Vision to Event Cameras", with
@supitalp
, Rene Ranftl, Vladlen Koltun.
PDF:
YouTube:
#eventcamera
#DVS
#deeplearning
#comnputervision
#UZH
#Intel
0
25
108
This week marks the 10th anniversary of my lab. This video celebrates the over 300 people who contributed to our research as Bsc/Msc/Ph.D. students, postdocs, visiting researchers, and collaborators! Enjoy:
6
7
110
2
23
108
We are hiring! We have multiple PhD and Postdoc openings in control, planning, aerodynamics, optimization, comp. vision, learning to contribute to the areas of:
- Autonomous drone racing,
- Autonomous inspection of power lines,
- Computational photography.
3
46
108
Congratulations to NASA and all their collaborators! First powered autonomous flight on another planet successful! Btw, it is also a vision-based flight (single-camera + IMU + laser altimeter).
2
14
106
Congratulations to my student
@DanielGehrig6
for successfully defending his Ph.D. in “Efficient, Data-Driven Perception with Event Cameras”! Many thanks to the external reviewers Marc Pollefeys
@mapo1
, Kostas Daniilidis
@KostasPenn
, and Andreas Geiger! Daniel has contributed
3
8
102
Join our annual workshop on “Integrated Perception, Learning, and Control for Agile Super Vehicles”, next Friday October 1 from 3pm Zurich time. We will have NASA/JPL, the head of the winning team of the DARPA SubT challenge and many more!
2
24
99
We have multiple openings for Phd students and Postdocs in machine learning for computer vision and vision-based robot navigation. Job descriptions and how to apply:
4
21
96
New self-folding drone squeezing through gaps, flattening against walls for up-close views, and grabbing objects with rotor arms. Great collaboration between
@UZH_ch
and
@EPFL_en
Paper:
thanks to
@nccrrobotics
1
40
96
Congratulations to my students Davide Falanga and Kevin Kleber for winning the NASA Tech Briefs Award out of 700 participants worldwide for the invention of the Foldable Drone [Falanga, RAL'19]!
#UZH
#Foldabledrone
#drones
0
19
94
That’s my lab :) Very proud of my guys! They will take Skydio drones to the next level ;-)
@davidefalanga
@foehnph
@kaufmann_elia
Today we're announcing the expansion of our world-class autonomy team with the hiring of four top PhD graduates from the Robotics and Perception Group. READ:
1
6
53
3
2
93
Check out our
#ICRA2024
paper, “Contrastive Learning for Enhancing Robust Scene Transfer in Vision-based Agile Flight,” where we present an approach to train agile vision-based flight policies that can adapt to and transfer across multiple unseen environments. We propose an
1
17
88
My former PhD student Antonio Loquercio, currently postdoc at Berkeley, will start working as assistant professor at the University of Pennsylvania’s GRASP Lab in Fall 2024! He is looking tor PhD students! Congratulations, Antonio, I’m very proud of you!
Excited to share that I'll join UPenn as an Assistant Professor next fall! I couldn't be more grateful to all the mentors and collaborators for their support along the way!
I'm looking for motivated students👩🎓to join my lab! If you love robots 🤖 and learning, look no further!
29
27
422
0
4
91
I am happy to share paper, dataset, code, video "Dronet: Learning to Fly by Driving", where we use Deep Neural Networks to teach drones to navigate autonomously in the streets of a city by imitating cars and bicycles, thus respecting traffic rules!
1
21
88
Apparently, Ingenuity flew over a texture-poor terrain and RANSAC wasn’t able to find inliers: According to NASA: “It's very featureless-bland, sandy terrain. And that's why we believe that during Flight 71, we had an emergency landing. She was flying over
Here's everything
@NASAJPL
knows about what likely happened on Ingenuity's last flight
#MarsHelicopter
0
3
21
1
15
85
Honored that our IEEE Transactions on Robotics 2020 paper "Deep Drone Racing: From Simulation to Reality with Domain Randomization" was selected Best Paper Award finalist! w/
@antoniloq
@kaufmann_elia
PDF, video, code:
1
3
89
DeepMind has acquired MuJoCo and is making it available for free. This is great news for robotics research!
We’ve acquired the MuJoCo physics simulator () and are making it free for all, to support research everywhere. MuJoCo is a fast, powerful, easy-to-use, and soon to be open-source simulation tool, designed for robotics research:
85
2K
6K
0
6
88
We are releasing
#EventScape
, a new dataset with
#events
, intensity
#frames
,
#semanticlabels
, and
#depthmaps
recorded in the
#CARLA
autonomous driving simulator!
Dataset & Paper:
Kudos to
@DanielGehrig6
@MathiasGehrig
@jhidalgocarrio
#eventcameras
0
20
88
Novel very promising localization technology by NASA/JPL. Works without GPS, RFID, or SLAM. Uses magnetoquasistatic fields. <1 m accuracy up to a few hundred meter range. Currently tested to localize firefighters.
A unique positioning technology, developed by
@NASAJPL
and
@DHSSciTech
, will help pinpoint firefighters inside buildings where other positioning technologies fail. The system, POINTER, is on track to be used by fire departments nationwide. Get the details:
28
207
2K
1
18
86
All the video recordings and slides of our
#IROS2020
workshop on Perception, Learning, and Control for Autonomous Agile Vehicles are online!
@loiannog
,
@SertacKaraman
1
14
87
Did you ever try to train your
#eventcamera
-based neural network with still images alone? It is now possible by using our new
#ECCV2022
paper! New semantic segmentation dataset is included!
Paper, datasets, code:
and
@Messikommer1
0
21
85
Only 5 days remaining to the
#SwissDroneDays
: June, 11-12, 2022, Zurich! We will attempt a world record with an autonomous AI-powered
#drone
. No registration required. Free for all. Info:
#robotics
#machinelearning
#deeplearning
#computervision
0
21
85