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.
Additional Information
© 2020 IEEE. We would like to thank the USGS and NPS for providing data. This work was supported by NSFG RFP Grant No. 1745301, the views are those of the authors and do not necessarily reflect the views of the NSF. Compute provided by Microsoft AI for Earth and AWS.Attached Files
Submitted - 1904.05916.pdf
Files
Name | Size | Download all |
---|---|---|
md5:5ff2ef4b617df2b791eda25786856854
|
6.6 MB | Preview Download |
Additional details
- Eprint ID
- 103465
- Resolver ID
- CaltechAUTHORS:20200526-140034764
- DGE-1745301
- NSF Graduate Research Fellowship
- Microsoft AI for Earth
- Amazon Web Services
- Created
-
2020-05-26Created from EPrint's datestamp field
- Updated
-
2021-11-16Created from EPrint's last_modified field
- Caltech groups
- Division of Biology and Biological Engineering