AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Low-Shot Visual Recognition by Shrinking and Hallucinating Features

Bharath Hariharan, Ross Girshick

ICCV 2017 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Transductive and semantic-dictionary zero-shot · 843 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · ImageNet unseen classes, 10 per classtop-1 accuracy: 29.79 -> 30.83Improved
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · ImageNet unseen classes, 2 per classtop-5 accuracy: 34.45 -> 38.76Improved
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · ImageNet unseen classes, 20 per classtop-1 accuracy: 35.75 -> 35.10Hurt
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · ImageNet unseen classes, all examplestop-1 accuracy: 51.41 -> 50.14Hurt
train with Siamese loss instead of gradient-norm loss10-layer ResNet · ImageNet unseen classes, 1 per classtop-1 accuracy: 9.80 -> 7.20Hurt
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · Birds, 1 per classtop-1 accuracy: 11.35 -> 13.50Improved
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · Aircraft, 5 per classtop-1 accuracy: 16.78 -> 16.76No clear effect
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · Aircraft, all examplestop-1 accuracy: 37.95 -> 35.07Hurt
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · Oxford flowers, 1 per classtop-1 accuracy: 42.95 -> 38.90Hurt
add squared gradient-norm penalty on small-batch classifier loss10-layer ResNet · VOC 2007 classification, 5 per classaccuracy: 51.94 -> 56.33Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy