AI Paper Atlas

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

Matching Visual Features to Hierarchical Semantic Topics for Image Paragraph Captioning

Dandan Guo, Ruiying Lu, Bo Chen, Zequn Zeng, Mingyuan Zhou

IJCV 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Concept-guided image captioning · 10 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 VTCM hierarchical topics to LSTM-ATTLSTM-ATT · Stanford image-paragraphCIDEr: 20.11 -> 22.82Improved
add VTCM hierarchical topics to M2 TransformerM2 Transformer · Stanford image-paragraphCIDEr: 16.1 -> 26.15Improved
add VTCM topics to M2 TransformerM2 Transformer · Stanford image-paragraphMETEOR: 15.4 -> 16.88Improved
add self-critical training to VTCM-LSTMVTCM-LSTM · Stanford image-paragraphCIDEr: 22.82 -> 25.50Improved
increase topic layers from 1 to 3 (ablation)VTCM-LSTM · Stanford image-paragraphCIDEr: 19.10 -> 22.82Improved
increase topic layers from 1 to 3 (ablation)VTCM-Transformer · Stanford image-paragraphCIDEr: 22.71 -> 26.15Improved
use variational topic encoder (VTCM) instead of two-stage PGBN (ablation)VTCM-LSTM · Stanford image-paragraphCIDEr: 20.49 -> 22.82Improved
use variational topic encoder (VTCM) instead of two-stage PGBN (ablation)VTCM-Transformer · Stanford image-paragraphCIDEr: 24.83 -> 26.15Improved
use multi-layer VTCM instead of ProdLDA shallow topics (ablation)LSTM LM · Stanford image-paragraphCIDEr: 18.52 -> 22.82Improved
use VTCM coupling instead of concatenating topics before softmax (ablation)LSTM LM · Stanford image-paragraphCIDEr: 18.02 -> 22.82Improved
use VTCM coupling instead of concatenating topics before softmax (ablation)M2 Transformer · Stanford image-paragraphCIDEr: 23.45 -> 26.15Improved
replace M2 Transformer with pretrained original Transformer (ablation)VTCM-Transformer · Stanford image-paragraphoverallHurt

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