AI Paper Atlas

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

Affective Image Content Analysis: Two Decades Review and New Perspectives

Sicheng Zhao, Xingxu Yao, Jufeng Yang, Guoli Jia, Guiguang Ding, Tat‐Seng Chua, Björn Wolfgang Schuller, Kurt Keutzer

TPAMI 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 212 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 global features to local features (ablation)WSCNet · FI2accuracy: 0.894 -> 0.896Improved
add global features to local features (ablation)PDANet · FI2accuracy: 0.807 -> 0.878Improved
add global features to local features (ablation)PDANet · Comicsaccuracy: 0.263 -> 0.588Improved
use global instead of local features (ablation)PDANet · FI8accuracy: 0.606 -> 0.696Improved
add local features to global features (ablation)PDANet · FI8accuracy: 0.696 -> 0.694No clear effect
add local features to global features (ablation)WSCNet · Instagramaccuracy: 0.816 -> 0.804Hurt
add local features to global features (ablation)WSCNet · Twitter IIaccuracy: 0.824 -> 0.815Hurt
use deep MVSO features instead of hand-crafted PAEFSVM/kNN/NB average · FI2accuracy: 0.717 -> 0.778Improved
use pre-trained VGGNet-16 features instead of PAEFSVM/kNN/NB average · FI8accuracy: 0.309 -> 0.561Improved
use high-level SentiBank instead of mid-level PAEFSVM/kNN/NB average · FI8accuracy: 0.309 -> 0.413Improved
use SVM instead of kNN classifierfeature classifiers · FI2accuracy: 0.721 -> 0.793Improved
use SVM instead of Naive Bayes classifierfeature classifiers · FI8accuracy: 0.349 -> 0.490Improved

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