AI Paper Atlas

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

BAM! The Behance Artistic Media Dataset for Recognition Beyond Photography

Michael J. Wilber, Chen Fang, Hailin Jin, Aaron Hertzmann, John Collomosse, Serge Belongie

ICCV 2017 · Vision-language and multimodal › Human and embodied modeling · approach: Human perception modeling · 165 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 BAM art labels instead of VOC/ImageNetobject classifier · BAM VOC categories averageAP: 0.9448 vs 0.8770 (YOLO), 0.8801 (SSD), 0.8567 (RN50)Improved
fuse ImageNet and BAM scores linearlyfine-tuned ResNet-50 · BAM VOC categories averageAP: 0.9448 -> 0.9512Improved
train jointly on BAM attributes (JAM)ResNet-50 · Flickr styleaccuracy: 0.376 -> 0.389Improved
train jointly on BAM attributes (JAM)ResNet-50 · Wikipaintings styleaccuracy: 0.505 -> 0.508No clear effect
train jointly on BAM attributes (JAM)ResNet-50 · AVA styleaccuracy: 0.603 -> 0.615Improved
train on seen media instead of unseen mediumobject classifier · Cross-medium object recognition (watercolor)accuracy: 63 -> 68Improved
train on 6 other media instead of ImageNet onlyobject classifier · Test on 3Daccuracy: 52 -> 70Improved
use StyleNet features instead of ImageNet ResNet featureslinear SVM · BAM emotion and media attributesAP: 4/4 emotion and 4/6 media attributesImproved
use StyleNet features instead of ImageNet ResNet featureslinear SVM · BAM content attributesAP: higher AP for ImageNet on all nine content attributesHurt
use task-specific subspaces instead of shared featuresretrieval embedding · BAM retrieval precision-at-10 (media)precision-at-10: 0.80 -> 0.91Improved
use task-specific subspaces instead of shared featuresretrieval embedding · BAM retrieval precision-at-10 (emotion)precision-at-10: 0.80 -> 0.84Improved
use task-specific subspaces instead of shared featuresretrieval embedding · BAM retrieval precision-at-10 (content)precision-at-10: 0.80 -> 0.71Hurt

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