AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

A Deep Matrix Factorization Method for Learning Attribute Representations

George Trigeorgis, Konstantinos Bousmalis, Stefanos P. Zafeiriou, Björn Wolfgang Schuller

TPAMI 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 367 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 depth (second factorization layer) to Semi-NMFSemi-NMF · CMU Multi-PIE identity classificationaccuracy: 36.46 -> 61.22Improved
add depth to Semi-NMFSemi-NMF · CMU Multi-PIE expression classificationaccuracy: 81.50 -> 80.54No clear effect
add attribute supervision and depth (Deep WSF) to Deep Semi-NMFDeep Semi-NMF · CMU Multi-PIE identity classificationaccuracy: 61.22 -> 65.17Improved
add depth to WSF (Deep WSF vs WSF)WSF · CMU Multi-PIE identity classificationaccuracy: 63.81 -> 65.17Improved
use all attributes (WSF-MA) instead of one attribute (WSF)WSF · CMU Multi-PIE identity classificationaccuracy: 63.81 -> 64.08Improved
add depth to Semi-NMFSemi-NMF · CMU PIE reconstruction (70 components)reconstruction error: 4.13 vs. 4.39No clear effect
add supervised pre-training on XM2VTS before unsupervised fine-tuning (ablation)Deep Semi-NMF · CMU PIE clusteringaccuracyImproved
use IGO features instead of pixel intensitiesDeep Semi-NMF · XM2VTS / CMU PIE clusteringaccuracyImproved

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