AI Paper Atlas

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

Tree Structure-Aware Few-Shot Image Classification via Hierarchical Aggregation

Min Zhang, Siteng Huang, Wenbin Li, Donglin Wang

ECCV 2022 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Alignment-based few-shot metric learning

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 HTS (tree-structured pretext aggregation, SSL variant)ProtoNet Conv4-64 · miniImageNet 5-way 1-shotaccuracy (%): 44.42 -> 58.96Improved
add HTS (SSL variant)ProtoNet Conv4-64 · miniImageNet 5-way 5-shotaccuracy (%): 64.24 -> 75.17Improved
add HTS (SSL variant)ProtoNet Conv4-64 · CUB-200-2011 5-way 1-shotaccuracy (%): 51.31 -> 67.32Improved
add HTS (SSL variant)GNN Conv4-64 · tieredImageNet 5-way 5-shotaccuracy (%): 55.31 -> 70.42Improved
add HTS (SSL variant)MatchingNet Conv4-64 · miniImageNet 5-way 1-shotaccuracy (%): 48.14 -> 52.29Mixed
use separate label spaces (HTS_SSL) instead of single label space (HTS_DA) (ablation)ProtoNet+HTS Conv4-64 · miniImageNet 5-way 1-shotaccuracy (%): 57.39 -> 58.96Improved
use separate label spaces (HTS_SSL) instead of single label space (HTS_DA) (ablation)MatchingNet+HTS Conv4-64 · miniImageNet 5-way 1-shotaccuracy (%): 53.64 -> 52.29Hurt
add HTS across domainsProtoNet Conv4-64 · tieredImageNet->miniImageNet 5-way 1-shotaccuracy (%): 47.01 -> 55.29Improved
add HTS across domainsProtoNet Conv4-64 · miniImageNet->CUB-200-2011 5-way 1-shotaccuracy (%): 32.18 -> 39.57Improved
add HTS across domainsProtoNet Conv4-64 · tieredImageNet->CUB-200-2011 5-way 5-shotaccuracy (%): 56.58 -> 60.71Improved
add plain pretext-task data augmentation (single label space, ERM) without HTSProtoNet · miniImageNet, CUB, CIFAR-FS, tieredImageNet 5-shotaccuracy (%)Hurt
add plain pretext-task self-supervised auxiliary losses without HTSProtoNet · miniImageNet, CUB, CIFAR-FS, tieredImageNet 5-shotaccuracy (%)Mixed

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