Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Low-shot Visual Recognition by Shrinking and Hallucinating Features

About

Low-shot visual learning---the ability to recognize novel object categories from very few examples---is a hallmark of human visual intelligence. Existing machine learning approaches fail to generalize in the same way. To make progress on this foundational problem, we present a low-shot learning benchmark on complex images that mimics challenges faced by recognition systems in the wild. We then propose a) representation regularization techniques, and b) techniques to hallucinate additional training examples for data-starved classes. Together, our methods improve the effectiveness of convolutional networks in low-shot learning, improving the one-shot accuracy on novel classes by 2.3x on the challenging ImageNet dataset.

Bharath Hariharan, Ross Girshick• 2016

Related benchmarks

TaskDatasetResultRank
Image ClassificationImageNet 1K Challenge (novel classes)
Top-5 Acc78.5
110
Few-shot classificationCUB
Accuracy81.1
96
Generalized Few-Shot LearningImageNet 2012 (Novel classes)
Top-5 Accuracy76.5
70
Few-shot Image ClassificationImageNet FS (novel)
Top-5 Acc0.82
59
Low-shot Image ClassificationImageNet 1k (novel classes)
Top-5 Acc79.5
57
Generalized Few-Shot LearningImageNet All classes 2012
Top-5 Accuracy78.5
50
Generalized Few-Shot LearningAWA2
Accuracy87.8
48
Few-shot Image ClassificationImageNet FS (all)
Top-5 Acc83.6
44
Low-shot Image ClassificationImageNet-1k (val)
Top-5 Accuracy85.2
40
Image ClassificationImagenet FS (All classes)
Top-5 Acc77.5
30
Showing 10 of 22 rows

Other info

Code

Follow for update