# Few-Shot Learning with Localization in Realistic Settings

arXiv: https://arxiv.org/abs/1904.08502 · HTML: https://mlsanity.com/paper/1904.08502/

## Key facts

- Cited by 34 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Apr 2019.

## Built on by (most cited first, 30 of 34)

- [BABEL: Bodies, Action and Behavior with English Labels](https://mlsanity.com/paper/2106.09696/) (2021)
- [Joint Distribution Matters: Deep Brownian Distance Covariance for Few-Shot Classification](https://mlsanity.com/paper/2204.04567/) (2022)
- [Danish Fungi 2020 -- Not Just Another Image Recognition Dataset](https://mlsanity.com/paper/2103.10107/) (2021)
- [Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-shot Learning](https://mlsanity.com/paper/2203.09064/) (2022)
- [Matching Feature Sets for Few-Shot Image Classification](https://mlsanity.com/paper/2204.00949/) (2022)
- [Task Discrepancy Maximization for Fine-grained Few-Shot Classification](https://mlsanity.com/paper/2207.01376/) (2022)
- [Context-Aware Meta-Learning](https://mlsanity.com/paper/2310.10971/) (2023)
- [Few-Shot Learning with Class Imbalance](https://mlsanity.com/paper/2101.02523/) (2021)
- [Learning to Affiliate: Mutual Centralized Learning for Few-shot Classification](https://mlsanity.com/paper/2106.05517/) (2021)
- [POODLE: Improving Few-shot Learning via Penalizing Out-of-Distribution Samples](https://mlsanity.com/paper/2206.04679/) (2022)
- [Decomposing Convolutional Neural Networks into Reusable and Replaceable Modules](https://mlsanity.com/paper/2110.07720/) (2021)
- [CancerGPT: Few-shot Drug Pair Synergy Prediction using Large Pre-trained Language Models](https://mlsanity.com/paper/2304.10946/) (2023)
- [DMN4: Few-shot Learning via Discriminative Mutual Nearest Neighbor Neural Network](https://mlsanity.com/paper/2103.08160/) (2021)
- [Towards Practical Few-Shot Query Sets: Transductive Minimum Description Length Inference](https://mlsanity.com/paper/2210.14545/) (2022)
- [Local Propagation for Few-Shot Learning](https://mlsanity.com/paper/2101.01480/) (2021)
- [Task-Oriented Channel Attention for Fine-Grained Few-Shot Classification](https://mlsanity.com/paper/2308.00093/) (2023)
- [Streaming Self-Training via Domain-Agnostic Unlabeled Images](https://mlsanity.com/paper/2104.03309/) (2021)
- [Unsupervised Meta-Learning via In-Context Learning](https://mlsanity.com/paper/2405.16124/) (2024)
- [Few-shot Classification with Hypersphere Modeling of Prototypes](https://mlsanity.com/paper/2211.05319/) (2022)
- [An Adaptive Plug-and-Play Network for Few-Shot Learning](https://mlsanity.com/paper/2302.09326/) (2023)
- [FILM: How can Few-Shot Image Classification Benefit from Pre-Trained Language Models?](https://mlsanity.com/paper/2307.04114/) (2023)
- [Adaptive receptive field-based spatial-frequency feature reconstruction network for fine-grained few-shot image classification](https://mlsanity.com/paper/2604.16936/) (2026)
- [Compositional Fine-Grained Low-Shot Learning](https://mlsanity.com/paper/2105.10438/) (2021)
- [Mutual-Information Based Few-Shot Classification](https://mlsanity.com/paper/2106.12252/) (2021)
- [Boosting Few-Shot Learning with Disentangled Self-Supervised Learning and Meta-Learning for Medical Image Classification](https://mlsanity.com/paper/2403.17530/) (2024)
- [Universal Algorithm-Implicit Learning](https://mlsanity.com/paper/2602.14761/) (2026)
- [Prototypical Region Proposal Networks for Few-Shot Localization and Classification](https://mlsanity.com/paper/2104.03496/) (2021)
- [Diagnosing and Remedying Shot Sensitivity with Cosine Few-Shot Learners](https://mlsanity.com/paper/2207.03398/) (2022)
- [Not All Instances Contribute Equally: Instance-adaptive Class Representation Learning for Few-Shot Visual Recognition](https://mlsanity.com/paper/2209.03034/) (2022)
- [Meta-Learning Transformers to Improve In-Context Generalization](https://mlsanity.com/paper/2507.05019/) (2025)

## Abstract (arXiv)

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.