CV4Smalls
Computer Vision with Small Data: A focus on Infants, Toddlers, and the Elderly

Workshop Overview
The CV4Smalls workshop is dedicated to exploring the challenges and innovations in applying computer vision techniques to scenarios where data is limited, with a specific emphasis on infants, toddlers, and the elderly. These domains present unique data constraints, and the workshop aims to foster discussions and solutions that address these challenges, leveraging computer vision to drive advancements in data-critical applications and improve quality of care and life for these populations.
Join us Virtually on February 28 at 9 AM MST 🖥️
Can’t make it to WACV in-person this year? No problem! We are pleased to announce that our workshop will also be live-streamed online at this link on zoom. Please join us on Friday, February 28th at 9am MST for a full day of exciting keynotes and paper presentations.
Workshop Sponsors
We are grateful to our workshop partners and sponsors Voxel51 and Tahmasbi Saki Foundation for Scientific Research for supporting CV4Smalls!
Workshop Timeline
The submission period for this workshop is now over.


Keynote Speakers

Babak Taati
Dr. Taati is a Senior Scientist at KITE, the research arm of the Toronto Rehabilitation Institute, and an Associate Professor at the University of Toronto, with affiliations at the Vector Institute, Intelligent Assistive Technology and Systems Lab (IATSL), and Computational Vision group. As the lead of KITE’s Aging team, he focuses on advancing assistive technologies for aging populations.
Talk Title: Computer Vision in Geriatric Care
As the global population ages, innovative technologies are crucial for providing quality care to individuals with physical or cognitive disabilities. This presentation explores cutting-edge computer vision systems designed to support and enhance elder care. We will discuss two applications: an ambient monitoring system for pain assessment in older adults with severe dementia, addressing the challenge of pain management in non-communicative patients, and a vision-based fall risk monitoring system for long-term care facilities, aimed at preventing falls and improving safety. These examples demonstrate the potential of computer vision to improve geriatric care, offering non-invasive, continuous monitoring solutions that can significantly improve quality of life and care efficiency for aging populations.

Satya Mallick
As the CEO of OpenCV.org, Dr. Satya Mallick leads the world’s largest open-source computer vision library, fostering its growth and success. Dr. Mallick is also the founder and partner of Big Vision LLC, a consulting company specializing in services and products across various domains, including computer vision, AI, deep learning, machine learning, ChatGPT & LLMs, augmented reality, and video analytics. Additionally, he runs LearnOpenCV.com, a widely recognized computer vision blog where he shares expertise and insights with a vast audience of readers.
Title: Small Data, Big Impact: Building Computer Vision Solutions for Underrepresented Populations.
In an era dominated by large-scale datasets, designing computer vision solutions for unique and often underrepresented populations demands rethinking technology and methodology. Infants, toddlers, and the elderly represent groups where data is scarce due to ethical, practical, and privacy concerns but where the potential for positive impact is immense. This keynote will explore the core challenges of building and deploying computer vision systems with limited yet precious data. We will see how approaches involving synthetic data, GenAI, stereo vision, and even models built using big data can immensely help in solutions where data is sparse, and privacy is critical.
Workshop Topics & Applications of Interest
Topics of Interest
- Transfer Learning, Domain adaptation
- Learning with Weak or No Supervision
- Zero-Shot and Few-Shot Learning
- Active and Continual Learning
- Learning with Data Augmentation: Data-Driven Generative Models
- Learning from Synthetic Data: Physics-Based Simulation
- Multi-Task and Meta Learning
- Learning with Enforced Structured Prior Knowledge
Applications of Interest
- Infants/Toddlers/Elderly Image Segmentation, Registration, and Classification
- Infants/Toddlers/Elderly 2D or 3D Pose Estimation, and Posture Classification
- Infants/Toddlers/Elderly Facial Landmark Detection and Tracking
- Infants/Toddlers/Elderly Action Recognition
- Infants/Toddlers/Elderly Multi-View Multi-Subject Tracking
- Infants/Toddlers/Elderly 3D Body Reconstruction, Image/Motion/Video Generation
- Infants/Toddlers/Elderly Behavior Analysis in Healthcare Applications