AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Contemplating Visual Emotions: Understanding and Overcoming Dataset Bias

Rameswar Panda, Jianming Zhang, Haoxiang Li, Joon-Young Lee, Xin Lu, Amit K. Roy-Chowdhury

ECCV 2018 · Vision-language and multimodal › Human and embodied modeling · approach: Human perception modeling · 88 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
train on large web WEBEmo data instead of Deep EmotionResNet-50 · Binary cross-dataset (mean others)accuracy (%): 63.52 -> 72.76Improved
train on WEBEmo instead of Emotion-6, tested on Emotion-6ResNet-50 · Emotion-6 testaccuracy (%): 77.72 -> 78.38Improved
add curriculum (2-6-25) instead of direct learning (ablation)ResNet-50 · Exploration study (mean others)accuracy (%): 68.13 -> 72.76Improved
add curriculum instead of self-directed learning (ablation)ResNet-50 · Exploration study (mean others)accuracy (%): 69.82 -> 72.76Improved
add curriculum instead of joint learning (ablation)ResNet-50 · Exploration study (mean others)accuracy (%): 70.38 -> 72.76Improved
use three-stage curriculum 2-6-25 instead of two-stage 2-6 (ablation)ResNet-50 · Self test and mean othersaccuracy (%): 78.21 -> 81.41 self; 70.05 -> 72.76 othersImproved
use 25 emotion categories instead of six for direct learning (ablation)ResNet-50 · Direct learning self testaccuracy (%): 2.31% drop with six categoriesImproved
train on WEBEmo versus Deep Emotion (sadness negative bias)ResNet-50 · sadness vs non-sadness% drop self to others: 25.09 -> -0.73Improved
use WEBEmo curriculum features instead of ImageNet featuresResNet-50 · UnBiasedEmoaccuracy (%): 64.20 -> 74.27Improved
use WEBEmo curriculum features instead of Deep Emotion featuresResNet-50 · UnBiasedEmoaccuracy (%): 68.88 -> 74.27Improved
use curriculum features instead of ImageNet featuresResNet-50 · Image Advertisementaccuracy (%): 23.42 -> 27.96Improved
use curriculum features instead of joint learning (ablation)ResNet-50 · VideoStory-P14accuracy (%): 47.25 -> 49.22Improved

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