AI Paper Atlas

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

Modeling Inter and Intra-Class Relations in the Triplet Loss for Zero-Shot Learning

Yannick Le Cacheux, Herve Le Borgne, Michel Crucianu

ICCV 2019 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Generative feature synthesis for zero-shot · 114 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 partial normalization of projected visual features (ablation)triplet loss ZSL model (theta+phi) · CUB ZSLper-class accuracy (%): 57.6 -> 63.8Improved
add flexible semantic margin (ablation)triplet loss ZSL model (theta+phi) · CUB ZSLper-class accuracy (%): 61.8 -> 63.8Improved
add relevance weighting of training samples (ablation)triplet loss ZSL model (theta+phi) · CUB ZSLper-class accuracy (%): 61.7 -> 63.8Improved
compute class distances with Mahalanobis distance (attribute correlations) rather than Euclidean (ablation)flexible semantic margin · CUB ZSLper-class accuracy (%): 61.7 -> 63.8Improved
add all three components (FSM, PN, RW) to a DEVISE-like triplet loss (ablation)triplet loss ZSL model (theta+phi) · CUB ZSLper-class accuracy (%): 56.6 -> 63.8Improved
add all three components (FSM, PN, RW) (ablation)triplet loss ZSL model (theta only) · CUB ZSLper-class accuracy (%): 55.0 -> 61.3Improved
learn projection of class prototypes too (theta+phi instead of theta only) (ablation)full model · CUB ZSLper-class accuracy (%): 61.3 -> 63.8Improved
apply calibration to penalize seen classes at predictionfull model · CUB GZSLharmonic mean H (%): 41.2 -> 53.0Improved
apply calibration to penalize seen classes at predictionfull model · GZSL average over datasetsharmonic mean H (%): 32.3 -> 50.4Improved
combine relevance weighting and flexible margin (ablation)triplet loss ZSL model · CUB ZSLper-class accuracy (%): 56.6 -> 61.0Improved

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