A Broader Study of Cross-Domain Few-Shot Learning

By Yunhui Guo, Noel C. F. Codella, Leonid Karlinsky, James Codella, John R. Smith, Kate Saenko, Tajana Rosing, Rogério Schmidt Feris. 2019

423 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Margin-Based Few-Shot Class-Incremental Learning with Class-Level Overfitting Mitigation (2022, background): “However, the model’s generalization capability on the downstream novel classes is much less explored and still needs to be improved [20, 14].”
Meta-Learning in Neural Networks: A Survey (2020, background): “Meanwhile, [222] challenges methods to generalize from the everyday ImageNet images to medical, satellite and agricultural images.”
How Well Do Self-Supervised Models Transfer? (2020, methodology): “Additionally, we evaluate on the Broader Study of Cross-Domain Few-Shot Learning (CD-FSL) benchmark introduced by [19].”
Revisiting Mid-Level Patterns for Cross-Domain Few-Shot Recognition (2020, neutral): “The novel-class domains may vary from being close to being distant against the base-class domain [14], because no assumptions could be made about what novel classes would appear and no one could enumerate all possible classes in base classes.”
Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need? (2020, background): “Nevertheless, despite the achieved improvements, several recent image classification works [2, 4, 6, 12, 32, 44] observed that meta-learning might have limited generalization capacity beyond the standard 1- or 5-shot classification benchmarks.”
TinyViT: Fast Pretraining Distillation for Small Vision Transformers (2022, methodology): “We also evaluate the transferability of TinyViT with different training settings on few-shot learning benchmark [25].”
A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, methodology): “Except for the Meta-Dataset [21] and BSCD-FSL [22], all other benchmark datasets were evaluated using 5-way-1-shot.”