AI4Space 2022

2nd Workshop on AI for Space

In conjunction with ECCV 2022

Date: 23 October 2022

Workshop attendance mode has changed to fully virtual
Accepted Papers
We thank our generous sponsors Spire Global and Draper Labs


The space sector is experiencing significant growth. Currently planned activities and utilisation models also greatly exceed the scope, ambition and/or commercial value of space missions in the previous century, e.g., autonomous spacecraft, space mining, and understanding the universe. Achieving these ambitious goals requires surmounting non-trivial technical obstacles. AI4Space focuses on the role of AI, particularly computer vision and machine learning, in helping to solve those technical hurdles. The workshop will highlight the space capabilities that draw from and/or overlap significantly with vision and learning research, outline the unique difficulties presented by space applications to vision and learning, and discuss recent advances towards overcoming those obstacles.


We are excited to co-host an AI for space challenge SPARK2022 with the workshop. The competition requires participants to develop data-driven approaches for spacecraft detection and trajectory estimation. SPARK will utilise data synthetically simulated with a state-of-the-art rendering engine in addition to data collected from the Zero-G Lab at the University of Luxembourg.

Final results and awards will be announced at AI4Space at ECCV 2022. Challenge winners may also be invited to present their methods at the workshop (subject to time availability in the finalised workshop program).

To the challenge!

Workshop Program

Block A (download calendar to add program to your calendar)

Tel Aviv (IDT=UTC+3)
23 October 2022
Los Angeles (PDT=UTC-7)
22/23 October 2022
Sydney (AEST=UTC+10)
23 October 2022
09:00 23:00 16:00 Welcome address + housekeeping
09:05 23:05 16:05 Keynote 1: Michal Segal-Rozenhaimer
09:29 23:29 16:29 Keynote 2: Simone D'Amico
09:53 23:53 16:53 Spotlight presentations 1
10:13 00:13 17:13 SPARK challenge + Keynote 3: Thanos Athanasiadis
10:38 00:38 17:38 Spotlight presentations 2
10:53 00:53 17:53 Breakout sessions
11:13 01:13 18:13 Award ceremony
11:18 01:18 18:18 End

Block B (download calendar to add program to your calendar)

Tel Aviv (IDT=UTC+3)
23 October 2022
Los Angeles (PDT=UTC-7)
23 October 2022
Sydney (AEST=UTC+10)
24 October 2022
19:00 09:00 02:00 Welcome address + housekeeping
19:05 09:05 02:05 Keynote 4: Lukas Mandrake
19:29 09:29 02:29 Keynote 5: Grzegorz Kakareko
19:53 09:53 02:53 Spotlight presentations 3
20:23 10:23 03:23 Breakout session
20:43 10:43 03:43 Award ceremony and close
20:48 10:48 03:48 End


Keynote Speakers

Michal Segal-Rozenhaimer
Tel Aviv University
Grzegorz Kakareko
Spire Global
Lukas Mandrake

Simone D'Amico
Stanford University

Accepted Papers

Spotlight Presentations 1 (Block A)

Order of presentation Paper title and authors
Transfer Learning for On-Orbit Ship Segmentation
Vincenzo Fanizza (Ubotica Technologies)*; David Rijlaarsdam (Ubotica Technologies); Pablo Tomás Toledano González (Ubotica Technologies); José Luis Espinosa-Aranda (Ubotica Technologies)
Spacecraft Pose Estimation Based on Unsupervised Domain Adaptation and on a 3D-Guided Loss Combination
Juan Ignacio Bravo (Deimos Space, Universidad Autónoma de Madrid)*; Alvaro Garcia-Martin (Universidad Autónoma de Madrid); Jesus Bescos (Universidad Autónoma de Madrid)
Monocular 6-DoF Pose Estimation for Non-cooperative Spacecrafts Using Riemannian Regression Network
Sunhao Chu (Shanghai Jiao Tong University)*; Shufan Wu (Shanghai Jiao Tong University); Yuxiao Duan (Shanghai Jiao Tong University); Klaus Schilling ()
Asynchronous Kalman Filter for Event-Based Star Tracking
Yonhon Ng (Australian National University)*; Yasir Latif (The University of Adelaide); Tat-Jun Chin (The University of Adelaide); Robert Mahony (Australian National University)

Spotlight Presentations 2 (Block A)

Order of presentation Paper title and authors
CubeSat-CDT: A Cross-Domain Dataset for 6-DoF Trajectory Estimation of a Symmetric Spacecraft
Mohamed Adel Musallam (SnT, University of Luxembourg)*; Arunkumar Rathinam (University of Luxembourg); Vincent Gaudilliere (SnT, University of Luxembourg); Miguel Ortiz del Castillo (SnT, University of Luxembourg); Djamila Aouada (SnT, University of Luxembourg)
End-to-end Neural Estimation of Spacecraft Pose with Intermediate Detection of Keypoints
Antoine Legrand (Université Catholique de Louvain)*; Renaud Detry (Katholieke Universiteit Leuven); Christophe De Vleeschouwer (Université Catholique de Louvain)
Globally Optimal Event-Based Divergence Estimation for Ventral Landing
Sofia A McLeod (University of Adelaide)*; Gabriele Meoni (European Space Agency); Dario Izzo (European Space Agency); Anne Mergy (European Space Agency); Daqi Liu (University of Adelaide); Yasir Latif (The University of Adelaide); Ian Reid (University of Adelaide); Tat-Jun Chin (The University of Adelaide)

Spotlight Presentations 3 (Block B)

Order of presentation Paper title and authors
Data Lifecycle Management in Evolving Input Distributions for Learning-Based Aerospace Applications
Somrita Banerjee (Stanford University)*; Apoorva Sharma (Stanford University); Edward Schmerling (Stanford University); Max Spolaor (The Aerospace Corporation); Michael Nemerouf (The Aerospace Corporation); Marco Pavone (Stanford University)
Improving Contrastive Learning on Visually Homogeneous Mars Rover Images
Isaac R Ward (Jet Propulsion Laboratory, California Institute of Technology)*; Charles Moore (Mississippi State University); Kai Pak (Jet Propulsion Laboratory, California Institute of Technology); Jingdao Chen (Mississippi State University); Edwin Goh (Jet Propulsion Laboratory)
MaRF: Representing Mars As Neural Radiance Fields
Lorenzo Giusti (Sapienza University of Rome, NASA JPL)*; Josue Garcia (University of California, San Diego); Steven Cozine (University of California, San Diego); Darrick Suen (University of California, San Diego); Christina Nguyen (University of California, San Diego); Shahrouz R Alimo (NASA JPL)
Mixed-Domain Training Improves Multi-Mission Terrain Segmentation
Grace M Vincent (North Carolina State University)*; Alice Yepremyan (Jet Propulsion Laboratory); Jingdao Chen (Mississippi State University); Edwin Goh (Jet Propulsion Laboratory)
Strong Gravitational Lensing Parameter Estimation with Vision Transformer
Kuan-Wei Huang (Carnegie Mellon University); Geoff Chih-Fan Chen (University of California, Los Angeles); Po-Wen Chang (Ohio State University); Sheng-Chieh Lin (University of Kentucky); Chia Jung Hsu (Chalmers University of Technology); Vishal Thengane (Mohamed bin Zayed University of Artificial Intelligence);Joshua Yao-Yu Lin (University of Illinois at Urbana-Champaign)*
Using Moffat Profiles to Register Astronomical Images
Mason G Schuckman (University of Maryland, Baltimore County); Roy E Prouty (University of Maryland, Baltimore County); David R Chapman (University of Maryland, Baltimore County); Don Engel (University of Maryland, Baltimore County)*

Important Dates

All times/dates below are in Pacific Standard Time (PST).
Paper Submission Deadline 17 June 2022 15 July 2022 11:59pm 22 July 2022
Notification to Authors 11 July 2022 11:59pm 15 August 2022
Camera-Ready Deadline 17 July 2022 20 August 2022 11:59am 22 August 2022
Workshop Date 23 October 2022

Call for Papers

We solicit papers for AI4Space. Papers will be reviewed and accepted papers will be published in the proceedings of ECCV Workshops. Authors of accepted papers will also be invited to present at the workshop (in hybrid mode) at ECCV 2022.

The general emphasis of AI4Space is vision and learning algorithms in off-Earth environments, including in the orbital region, surface and underground environments on other planetary bodies (e.g., the moon, Mars and asteroids), interplanetary space and solar system, and distant galaxies. Target application areas include autonomous spacecraft, space robotics, space traffic management, astronomy, astrobiology and cosmology. Emphasis is also placed on novel sensors and processing hardware for vision and learning in space, mitigating the challenges of the space environment towards vision and learning (e.g., solar radiation, extreme temperatures), and solving practical difficulties in vision and learning for space (e.g., lack of training data, unknown or partially known characteristics of operating environments).

A specific list of topics is as follows:

  • Visual navigation for spacecraft operations (rendezvous, docking, maneuvers, entry descent landing)
  • Vision and learning for space robotics
  • GPS-denied positioning on the moon and Mars
  • Space debris monitoring and mitigation
  • Vision and learning for astronomy, astrobiology and cosmology
  • Novel sensors for space applications
  • Processing hardware for vision and learning in space
  • Mitigating challenges of the space environment to vision and learning
  • Datasets, transfer learning and domain gap for space problems

Peaceful usage of AI for space

All papers published via this workshop must be aimed towards the peaceful usage of AI for space.


Submissions should be in the ECCV format and are limited to 14 pages excluding reference. Use the template for detailed formatting instructions. Please refer to the main conference's submission guildelines for more details, and do not hesitate to contact the lead organiser Tat-Jun Chin if you have any questions.

Paper submission will be conducted through CMT3. Submission deadline

Reviewing is double blind - remember to remove your names and affiliations in the submitted version (selecting the reviewing option in the LaTeX template will take care of that). Accepted works will be published in the ECCV 2022 proceedings.


Tat-Jun Chin
The University of Adelaide
Luca Carlone
Massachusetts Institute of Technology
Djamila Aouada
University of Luxembourg
Binfeng Pan
Northwestern Polytechnical University
Viorela Ila
The University of Sydney
Benjamin Morrell
NASA Jet Propulsion Lab
Grzegorz Kakareko
Spire Global
Sofia Mcleod
The University of Adelaide

Program Committee

  • Abhishek Thakur, BRINC
  • Alan Liew, Griffith University
  • Alina Bialkowski, The University of Queensland
  • Andres Moya, Universidad Politécnica de Madrid
  • Andrzej Kucik, European Space Agency
  • Anirudh Chakravarthy, Carnegie Mellon University
  • Anis Kacem, SnT, University of Luxembourg
  • Artur Nowakowski, European Space Agency
  • Arunkumar Rathinam, University of Luxembourg
  • Benjamin Morrell, NASA JPL
  • Binfeng Pan, Northwestern Polytechnical University
  • Bo Chen, The University of Adelaide
  • Carl Seubert, SmartSat CRC
  • Chee Kheng Chng, The University of Adelaide
  • Daqi Liu, The University of Adelaide
  • David Suter, Edith Cowan University
  • Deegan Atha, NASA JPL
  • Diego Valsesia, Politecnico di Torino
  • Enjie Ghorbel, SnT, University of Luxembourg
  • Evridiki Ntagiou, European Space Agency
  • Florent Lafarge, INRIA
  • Gabriele Meoni, European Space Agency
  • Gary Doran, NASA JPL, Machine Learning and Instrument Autonomy
  • Giovanni Beltrame, Polytechnique Montreal
  • Giulia Ciabatti, Sapienza, University of Rome
  • Huangying Zhan, The University of Adelaide
  • Jacopo Villa, University of Colorado Boulder
  • Jessica Todd, Massachusetts Institute of Technology
  • Jian-Feng Shi, MDA
  • Joshua Critchley-Marrows, The University of Sydney
  • Kapil Mirchandani, Pune Institute of Computer Technology
  • Katherine Skinner, University of Michigan
  • Keenan Albee, Massachusetts Institute of Technology
  • Leo Pauly, SnT, University of Luxembourg
  • Manuel Salvoldi Ben-Gurion, University of the Negev
  • Marie Farrell, Maynooth University
  • Mark Rutten, In Track Solutions
  • Michele Sasdelli, The University of Adelaide
  • Mickaël Laîné, Space Robotics Lab., Tohoku University
  • Moritz Von Looz, European Space Agency
  • Padmaja Jonnalagedda, UC Riverside
  • R. Michael Swan, NASA JPL
  • Roberto Lopez-Sastre, University of Alcala
  • Roberto Del Prete, University of Naples Federico II
  • Roberto Capobianco, Sapienza University of Rome & Sony AI
  • Rodrigo Ventura, IST, Lisbon
  • Roshan Roy, Birla Institute of Technology and Science, Pilani
  • Shirley Ho, Flatiron Institute
  • Sk Aziz Ali, SnT, University of Luxembourg
  • Sriram Baireddy, Purdue University
  • Sumant Sharma, Stanford University
  • Sutharsan Mahendren, Queensland University of Technology
  • Vincent Gaudilliere, SnT, University of Luxembourg
  • Xue Wan, Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences Sciences
  • Yasir Latif, The University of Adelaide
  • Yuanbin Shao, Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences Sciences