AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Customized Image Narrative Generation via Interactive Visual Question Generation and Answering

Andrew Shin, Yoshitaka Ushiku, Tatsuya Harada

CVPR 2018 · Image generation › Autoregressive and latent generation · approach: Compositional scene generation · 6 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
Generate narratives using regional self-Q&A rather than standard image captioningImage narrative generation · MS COCO human evaluationOverall: 3.201 -> 3.391Improved
Generate narratives using regional self-Q&A rather than standard image captioningImage narrative generation · MS COCO test 2014BLEU1: 20.77 -> 20.87Improved
Generate narratives using regional self-Q&A rather than standard image captioningImage narrative generation · MS COCO test 2014BLEU4: 1.90 -> 1.41Hurt
Use VQG and VQA together to select questions with multiple plausible answersInteractive question generation · Interactive question evaluationvotes for multiple responses: 217 -> 664Improved
Include question context in preference learning and narrative generation (ablation)Same-image user choice reflection · Same-image user choice reflectionAvg. Score: 2.636$\pm$1.01 -> 3.851$\pm$1.12Improved
Use learned user preferences to customize narratives on new imagesNew-image interest reflection · New-image user interest reflectionAvg. Score: 2.238$\pm$1.24 -> 3.455$\pm$0.93Improved
Train preference learning with choices from the same user rather than random user matches (ablation)New-image preference learning · New-image choice predictionaverage score: 45.17 -> 68.72Improved
Train preference learning with randomly matched user choices rather than image features only (ablation)New-image interest reflection · New-image user interest reflectionAvg. Score: 2.238$\pm$1.24 -> 2.772$\pm$0.79Improved
Replace the regional question-answer approach with directly training on collected long narratives (ablation)Conventional image narrative captioning · MS COCO human-written narrativesdescriptive accuracyHurt
Use DenseCap instead of OursImage narrative generation · MS COCO human evaluationAccuracy: 3.132 -> 3.188Improved

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