Few-Shot Learning with Localization in Realistic Settings
About
Traditional recognition methods typically require large, artificially-balanced training classes, while few-shot learning methods are tested on artificially small ones. In contrast to both extremes, real world recognition problems exhibit heavy-tailed class distributions, with cluttered scenes and a mix of coarse and fine-grained class distinctions. We show that prior methods designed for few-shot learning do not work out of the box in these challenging conditions, based on a new "meta-iNat" benchmark. We introduce three parameter-free improvements: (a) better training procedures based on adapting cross-validation to meta-learning, (b) novel architectures that localize objects using limited bounding box annotations before classification, and (c) simple parameter-free expansions of the feature space based on bilinear pooling. Together, these improvements double the accuracy of state-of-the-art models on meta-iNat while generalizing to prior benchmarks, complex neural architectures, and settings with substantial domain shift.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Few-shot classification | tieredImageNet (test) | -- | 282 | |
| Few-shot classification | CUB (test) | -- | 145 | |
| Few-shot Image Classification | miniImageNet (test) | -- | 111 | |
| Few-shot Image Classification | mini-Cars (test) | Accuracy52.9 | 28 | |
| Few-shot classification | i-Nat (test) | Accuracy51.25 | 26 | |
| Image Classification | miniImageNet -> FGVC-Aircraft (test) | Accuracy63.56 | 8 |