AI Paper Atlas

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

Semi-transductive Learning for Generalized Zero-Shot Sketch-Based Image Retrieval

Ce Ge, Jingyu Wang, Qi Qi, Haifeng Sun, Tong Xu, Jianxin Liao

AAAI 2023 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Transductive and semantic-dictionary zero-shot · 13 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
Use semi-transductive learning with unlabeled unseen-class imagesSTL · SketchymAP (%): 44.5 -> 53.0Improved
Use semi-transductive learning with unlabeled unseen-class imagesSTL · TU-BerlinmAP (%): 31.2 -> 40.2Improved
Add discriminative semantic alignment to ranking learning (ablation)STL · SketchymAP (%): 16.3 -> 27.4Improved
Add discriminative semantic alignment to ranking learning (ablation)STL · TU-BerlinmAP (%): 9.6 -> 13.8Improved
Add distribution calibration using unlabeled unseen-class images and semi-transductive ranking (ablation)STL with semantic alignment · SketchymAP (%): 27.4 -> 53.0Improved
Add distribution calibration using unlabeled unseen-class images and semi-transductive ranking (ablation)STL with semantic alignment · TU-BerlinmAP (%): 13.8 -> 40.2Improved
Use a semi-heterogeneous encoder instead of a heterogeneous encoder (ablation)STL ranking model · SketchymAP (%): 8.7 -> 16.3Improved
Use a semi-heterogeneous encoder instead of a heterogeneous encoder (ablation)STL ranking model · TU-BerlinmAP (%): 5.3 -> 9.6Improved
Use a semi-heterogeneous encoder instead of a Siamese encoder (ablation)STL ranking model · SketchymAP (%): 11.2 -> 16.3Improved
Use a semi-heterogeneous encoder instead of a Siamese encoder (ablation)STL ranking model · TU-BerlinmAP (%): 6.7 -> 9.6Improved
Compress the embedding to 64D rather than use the 1024D STL embedding (ablation)STL · SketchymAP (%): 53.0 -> 44.6Hurt
Compress the embedding to 64D rather than use the 1024D STL embedding (ablation)STL · TU-BerlinmAP (%): 40.2 -> 36.3Hurt

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