Annotation-Efficient Object Detection: Unsupervised Discovery to Active Learning
Offered By: VinAI via YouTube
Course Description
Overview
Explore annotation-efficient approaches to object detection in this comprehensive seminar by AI Research Scientist Huy V. Vo from FAIR Labs, Meta. Delve into alternatives to fully-supervised object detection that require less or no manual annotation, including unsupervised object discovery and active learning strategies. Learn about optimization-based approaches, ranking methods, and seed-growing techniques that leverage self-supervised transformers for identifying and localizing similar objects in image collections without manual annotation. Discover the BiB active learning strategy, designed to combine weakly-supervised and active learning for training object detectors, offering an improved balance between annotation cost and effectiveness compared to traditional methods. Gain insights into cutting-edge research on learning problems with limited supervision, essential for developing intelligent systems like autonomous vehicles and robots.
Syllabus
VinAI Seminar Huy Vo Annotation Efficient
Taught by
VinAI
Related Courses
Artificial Intelligence Foundations: Thinking MachinesLinkedIn Learning Deep Learning for Computer Vision
NPTEL via YouTube NYU Deep Learning
YouTube Stanford Seminar - Representation Learning for Autonomous Robots, Anima Anandkumar
Stanford University via YouTube A Path Towards Autonomous Machine Intelligence - Paper Explained
Yannic Kilcher via YouTube