Latent Embedding Feedback and Discriminative Features for Zero-Shot Classification

Open paper
Benchmark Model Rank Results
generalized-zero-shot-learning-on-awa2GZSL_TF-VAEGAN#2Harmonic mean: 66.6
generalized-zero-shot-learning-on-cub-200GZSL_TF-VAEGAN#1Harmonic mean: 58.1
generalized-zero-shot-learning-on-sunGZSL_TF-VAEGAN#1Harmonic mean: 43
zero-shot-learning-on-awa2ZSL_TF-VAEGAN#2average top-1 classification accuracy: 72.2
zero-shot-learning-on-cub-200-2011ZSL_TF-VAEGAN#2average top-1 classification accuracy: 64.9
zero-shot-learning-on-sun-attributeZSL_TF-VAEGAN#1average top-1 classification accuracy: 66

Rank counts only results with a code link.