AI Paper Atlas

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

CLASTER: Clustering with Reinforcement Learning for Zero-Shot Action Recognition

Shreyank N Gowda, Laura Sevilla-Lara, Frank Keller, Marcus Rohrbach

ECCV 2022 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Generative feature synthesis for zero-shot

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 clustering of representation (ablation)CLASTER ZSL · UCF101 ZSLaccuracy (%): 31.6 -> 46.4Improved
add clustering of representation (ablation)CLASTER ZSL · HMDB51 ZSLaccuracy (%): 25.6 -> 36.8Improved
use meaningful (K-means) clusters instead of random clusters (ablation)CLASTER ZSL · UCF101 ZSLaccuracy (%): 24.1 -> 35.3Improved
add random clustering (ablation)CLASTER ZSL · UCF101 ZSLaccuracy (%): 31.6 -> 24.1Hurt
add semantic embedding in the clustered representation (ablation)CLASTER ZSL · UCF101 ZSLaccuracy (%): 39.4 -> 46.4Improved
add semantic embedding in the clustered representation (ablation)CLASTER ZSL · HMDB51 ZSLaccuracy (%): 27.5 -> 36.8Improved
optimize centroids with RL instead of SGD (ablation)CLASTER ZSL · UCF101 ZSLaccuracy (%): 39.1 -> 46.4Improved
optimize centroids with RL instead of NetVLAD (ablation)CLASTER ZSL · UCF101 ZSLaccuracy (%): 41.7 -> 46.4Improved
optimize clusters with RL instead of K-means (ablation)CLASTER · UCF101 train puritypurity: 0.77 -> 0.89Improved
concatenate visual and semantic features instead of averaging (ablation)CLASTER ZSL · HMDB51 ZSL (ED)accuracy (%): 33.1 -> 43.2Improved
concatenate instead of dot product (ablation)CLASTER ZSL · HMDB51 ZSL (ED)accuracy (%): 33.9 -> 43.2Improved
combine word2vec and sen2vec embeddingsCLASTER · ZSL average over datasetsaccuracy (%): 0.3%, 0.8% and 0.9%Improved

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