Paper Info
Title
Synthetic Examples Improve Generalization for Rare Classes
Abstract
The ability to detect and classify rare occurrences in images has important applications - for example, counting rare and endangered species when studying biodiversity, or detecting infrequent traffic scenarios that pose a danger to self-driving cars. Few-shot learning is an open problem: current computer vision systems struggle to categorize objects they have seen only rarely during training, and collecting a sufficient number of training examples of rare events is often challenging and expensive, and sometimes outright impossible. We explore in depth an approach to this problem: complementing the few available training images with ad-hoc simulated data.Our testbed is animal species classification, which has a real-world long-tailed distribution. We present two natural world simulators, and analyze the effect of different axes of variation in simulation, such as pose, lighting, model, and simulation method, and we prescribe best practices for efficiently incorporating simulated data for real-world performance gain. Our experiments reveal that synthetic data can considerably reduce error rates for classes that are rare, that as the amount of simulated data is increased, accuracy on the target class improves, and that high variation of simulated data provides maximum performance gain.
Year
DOI
Venue
2020
10.1109/WACV45572.2020.9093570
2020 IEEE Winter Conference on Applications of Computer Vision (WACV)
Keywords
DocType
ISSN
biodiversity,few-shot learning,computer vision,ad-hoc simulated data,animal species classification,real-world long-tailed distribution,natural world simulators
Conference
2472-6737
ISBN
Citations 
PageRank 
978-1-7281-6554-7
0
0.34
References 
Authors
15
8
Name
Order
Citations
PageRank
Sara Beery1122.91
Yang Liu200.34
Dan Morris31691100.70
Jim Piavis400.34
Ashish Kapoor51833119.72
Markus Meister6142.07
Neel Joshi7115563.95
pietro perona8164331969.06