AI Paper Atlas

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

Describing Textures using Natural Language

Chenyun Wu, Mikayla Timm, Subhransu Maji

ECCV 2020 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure

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
use BERT instead of mean-pooled word embeddings as phrase encoder (ablation)MetricLearning · DTD2 validation phrase retrievalMAP: 18.80 -> 31.68Improved
use BERT instead of biLSTM as phrase encoder (ablation)MetricLearning · DTD2 validation phrase retrievalMAP: 23.53 -> 31.68Improved
use BERT instead of ELMo as phrase encoder (ablation)MetricLearning · DTD2 validation phrase retrievalMAP: 28.13 -> 31.68Improved
use BERT instead of mean-pooled word embeddings as phrase encoder (ablation)MetricLearning · DTD2 validation image retrieval from phraseMAP: 7.19 -> 15.22Improved
train metric learning on descriptions instead of phrasesMetricLearning · DTD2 image retrieval from descriptionMRR: 12.92 -> 13.95Improved
metric learning with BERT instead of classifierDTD2 baselines · DTD2 test phrase retrievalMAP: 27.12 -> 31.77Improved
metric learning with BERT instead of classifierDTD2 baselines · DTD2 test image retrieval from phraseMAP: 14.75 -> 13.50Hurt
metric-learning top-5 phrases instead of Show-Attend-Tell generationdescription generation · DTD2 description generationCIDEr: 37.47 -> 47.34Improved
use ResNet layers 2 and 4 features instead of layer 1 only (ablation)Classifier · DTD2 validation phrase retrievalMAP: 13.10 -> 26.57Improved
use metric learning on description instead of phrase for Color+Pattern queryMetricLearning · Synthetic textures Color+PatternR-Precision: 41.73 -> 35.77Hurt
models queried with foreground color of pattern vs random chanceMetricLearning - phrase · Synthetic textures ForegroundR-Precision: 46.55 vs random 50.00No clear effect
combine CUB attributes with DTD2 texture attributeslogistic regression on CUB · CUB 200-way classificationaccuracyImproved

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