2021

Held in conjunction with CVPR 2021 (Virtual), June 25
Main Theme: Embedded Vision for Social Good
Organized by: Marius Leordeanu, Anelia Angelova, Viorica Patraucean, Tse-Wei Chen

Description

Embedded vision is an active field of research, bringing together efficient learning models with fast computer vision algorithms, to tackle many areas of robotics and intelligent systems that are enjoying an impressive growth today. Such strong impact comes with many challenges that stem from the difficulty of understanding complex visual scenes under the tight computational constraints required by real-time solutions on embedded devices. These challenges impose two main directions: 1) the development of lightweight, but highly accurate and fast machine learning and vision models and 2) the creation of efficient hardware, appropriate for the models in 1.

We aim to bring together top researchers and practitioners who actively participate in these research directions. Our theme this year revolves around embedded vision for social good. We want to celebrate the strong positive impact that works in this field start having in different areas of the society, such as agriculture, environment, search and rescue, medicine, sign language, and others.

Program

Post your questions here!
https://app.sli.do/event/syrub90x

Time (GMT+1)	Session	Speaker	Topic	Moderators
15:00-15:15	Welcome notes	EVW Committee	Welcome notes	Marius Leordeanu, Tse-Wei Chen
15:15-15:45	Invited Talk #1	Richard Bowden	Building a Universal Sign Translator
15:45-16:15	Invited Talk #2	Joyce Nabende	Building and Deploying End-to-End AI and Data Systems in a Developing World
16:15-16:30	Break
16:30-17:00	Invited Talk #3	Andrew Davison	Representations and Computational Patterns for Spatial AI
17:00-17:15	Paper ID #1	Tse-Wei Chen	CASSOD-Net: Cascaded and Separable Structures of Dilated Convolutional Neural Networks for Embedded Vision Systems and Applications
17:15-17:30	Paper ID #4	Nima Aghli	Combining Weight Pruning and Knowledge Distillation for CNN Compression
17:30-17:45	Paper ID #5	Jamie Menjay Lin	Phase Selective Convolution
17:45-17:55	Demo #9	Mehdi Bahri	Binary Graph Neural Networks
17:55-18:05	Demo #13	Saad Abbasi	OutlierNets: Highly Compact Deep Autoencoder Network Architectures for On-Device Acoustic Anomaly Detection
18:05-19:15	Long Break
19:15-19:45	Invited Talk #4	Antonios Gasteratos	The need for embedded systems in airborne Search and Rescue missions	Anelia Angelova, Viorica Patraucean
19:45-21:15	Invited Talk #5	Kurt Keutzer	From Results Tables to Applications: Realizing the State-of-the-Art at the Edge
20:15-21:00	Break
21:00-21:15	Paper ID #11	Catalin Ristea	Automotive Radar Interference Mitigation with Unfolded Robust PCA based on Residual Overcomplete Auto-Encoder Blocks
21:15-21:30	Paper ID #15	Mihai Cristian Pîrvu	Depth distillation: unsupervised metric depth estimation for UAVs by finding consensus between kinematics, optical flow and deep learning
21:30-21:40	Demo ID #17	Sorin Grigorescu	Embedded Vision for Self-Driving on Forest Roads
21:40-21:50	Demo ID #12	Mehdi Bahri	Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation
21:50-22:00	Demo ID #10	Quoc-Huy Tran	RetroActivity: Rapidly Deployable Live Task Guidance Experiences
22:00-22:15	Closing Remarks

New!
All accepted papers are invited to submit extended versions to a Special Issue of Journal: Sensors (IF 3.275) on Efficient Sensing, Learning and Vision for Autonomous Robotics.
Guest Editor: Marius Leordeanu.
Deadline: (TBD)
The workshop papers accepted to the journal will receive a 30% discount for the Open Access fee.

Important Dates

Paper submission: April 2, 2021
Demo abstract submission: April 2, 2021
Notification to the authors: April 15, 2021
Camera ready paper: April 18, 2021

Due to numerous requests for an extension, we have decided to move the deadline of paper submission to April 2, 2021.

---

Best paper award sponsored by Google: $500

---

Invited Speakers

Andrew Davison
Title of Talk: Representations and Computational Patterns for Spatial AI
Biography: Andrew Davison is Professor of Robot Vision and Director of the Dyson Robotics Laboratory at Imperial College London. His long-term research focus is on SLAM (Simultaneous Localisation and Mapping) and its evolution towards general `Spatial AI’: computer vision algorithms which enable robots and other artificial devices to map, localise within and ultimately understand and interact with the 3D spaces around them. With his research group and collaborators he has consistently developed and demonstrated breakthrough systems, including MonoSLAM, KinectFusion, SLAM++ and CodeSLAM, and recent prizes include Best Paper at ECCV 2016 and Best Paper Honourable Mention at CVPR 2018. He has also had strong involvement in taking this technology into real applications, in particular through his work with Dyson on the design of the visual mapping system inside the Dyson 360 Eye robot vacuum cleaner and as co-founder of applied SLAM start-up SLAMcore. He was elected Fellow of the Royal Academy of Engineering in 2017.

---

Antonios Gasteratos
Title of Talk: The need for embedded systems in airborne Search and Rescue missions
Biography: Prof. Antonios Gasteratos (FIET) is a Full Professor of Robotics, Mechatronics and Computer Vision at Democritus University of Thrace, Deputy Head of the Department of Production and Management Engineering and Director of the Laboratory of Robotics and Automation. He holds a MEng. and a PhD in Electrical and Computer Engineering, Democritus University of Thrace (1994 and 1999, respectively). During the past 15 years he has been principal investigator to several projects funded mostly by the European Commission, the European Space Agency, the Greek Secretariat for Research and Technology, Industry and other sources, which were mostly related to robotics and vision. He has published over 200 papers in peer reviewed journals and international conferences and written 3 textbooks in Greek. He is Subject Editor-in-Chief in Electronics Letters and Assoc. Editor at the International Journal of Optomechatronics. He is also evaluator of projects supported by the European Commission and other funding agencies, as well as reviewer in many international journals in the field of Computer Vision and Robotics. Antonios Gasteratos has been a member of programme committees of international conferences and chairman and co-chairman of international conferences and workshops.

---

Joyce Nakatumba-Nabende
Title of Talk: Building and Deploying End-to-End AI and Data Systems in a Developing World
Biography:
Dr. Joyce NakatumbaNabende is a lecturer in the Department of Computer Science at Makerere University and the head of the Makerere Artificial Intelligence and Data Science research Lab. Joyce holds a PhD in Computer Science from Eindhoven University of Technology, The Netherlands. Currently, Joyce leads a team of researchers that carry out research in the application of machine learning techniques in solving problems in the areas of agriculture and health. She also carries out research in natural language processing and building data resources for under resourced Uganda languages.

---

Kurt Keutzer
Title of Talk: From Results Tables to Applications: Realizing the State-of-the-Art at the Edge
Biography: Kurt Keutzer is a Professor of EECS at University of California, Berkeley. His research covers all aspects of Deep Learning. His collaboration on the LARS and LAMB algorithms reduced the training time of ImageNet and BERT to minutes. His “Squeeze” family of DNNs were among the first DNNs suitable for mobile and IOT applications. His work on the HAWQ family of quantizers has brought a systematic approach to low-bit precision quantization. Recent work on I-BERT has created a new generation of DNNs that require no floating-point computation. Kurt’s research has applied these advances to bring computer vision, natural language understanding, multi-media analytics, and speech recognition to edge devices.  Previously, Kurt was CTO at Synopsys, and his contributions to Electronic Design Automation were recognized at the 50^th Design Automation Conference where he was noted as a Top 10 most cited author, author of a Top 10 cited paper, and one of only three people to win four Best Paper Awards in the 50 year history of that conference.  As an entrepreneur, Kurt has been an investor and advisor to over 25 startups. His most recent exits have been DeepScale (co-founder), acquired by Tesla, and BabbleLabs (investor and advisor), acquired by Cisco.

---

Richard Bowden
Title of Talk: Building a Universal Sign Translator
Biography:
Richard Bowden is a Professor at University of Surrey, and a leading expert in computer vision for sign language technologies.

Topics

• Ethics in embedded vision
• Applications of embedded vision in domains with high social impact, for example:
  • Agriculture, forestry, wildlife, environment
  • Search and rescue
  • Sign language
  • Medicine and healthcare
  • Education
• Applications of embedded vision systems in general domains: 
  • UAVs (industrial, mobile and consumer)
  • Advanced assistance systems and autonomous navigation frameworks
  • Augmented and Virtual Reality
  • Robotics
• New trends and challenges in embedded visual processing
• Analysis of vision problems specific to embedded systems
• Analysis of embedded systems issues specific to computer vision
• Lightweight and efficient computer vision algorithms for embedded systems
• Hardware dedicated to embedded vision systems (GPUs, FPGAs, DSPs, etc.)
• Software platforms for embedded vision systems
• Biologically-inspired vision and embedded systems
• Hardware and software enhancements that impact vision applications
• Performance metrics for evaluating embedded systems
• Hybrid embedded systems combining vision and other sensor modalities
• Embedded vision systems applied to new domains

Committee

Steering Committee:
Marilyn Claire Wolf, University of Nebraska-Lincoln
Martin Humenberger, NAVER LABS Europe
Roland Brockers, Jet Propulsion Laboratory
Swarup Medasani, MathWorks
Stefano Mattoccia, University of Bologna
Jagadeesh Sankaran, Nvidia
Goksel Dedeoglu, Perceptonic
Margrit Gelautz, Vienna University of Technology
Branislav Kisacanin, Nvidia
Sek Chai, Latent AI
Zoran Nikolic, Nvidia
Ravi Satzoda, Nauto
Stephan Weiss, University of Klagenfurt
Ahmed Nabil Belbachir, NORCE Norwegian Research Centre

Program Committee (Tentative):
Alina Marcu, University “Politehnica” of Bucharest
Antonio Haro, eBay
Burak Ozer, Verificon Corporation
Daniel Steininger, AIT Austrian Institute of Technology
Dongchao Wen, Canon Information Technology (Beijing) Co., LTD
Dragos Costea, University “Politehnica” of Bucharest
Faycal Bensaali, Qatar University
Hassan Rabah, University of Lorraine
Jeff Delaune, Jet Propulsion Laboratory, California Institute of Technology
Jörg Thiem, University of Applied Sciences and Arts Dortmund
Linda Wills, Georgia Institute of Technology
Martin Kampel, Vienna University of Technology, Computer Vision Lab
Matteo Poggi, University of Bologna
Matthias Schörghuber, AIT Austrian Institute of Technology
Muling Guo, Canon Inc.
Nabil Belbachir, Norwegian Research Centre AS
Sek Chai, Latent AI
Senthil Yogamani, Valeo Vision Systems
Stefano Mattoccia, University of Bologna
Stephan Weiss, Klagenfurt University of Austria
Supriya Sathyanarayana, Stanford University
Tse-Wei Chen, Canon Inc.