AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

In Defense of Scene Graphs for Image Captioning

Kien Nguyen, Subarna Tripathi, Bang Du, Tanaya Guha, Truong Q. Nguyen

ICCV 2021 · Graph learning › Knowledge and temporal graphs · approach: Scene graph generation · 77 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 bounding-box node grounding (ablation)SG2Caps baseline · COCO Karpathy test (ablation, cross-entropy)CIDEr: 92.6 -> 97.4Improved
add HOI graph on top of BBox (ablation)SG2Caps · COCO Karpathy test (ablation, cross-entropy)CIDEr: 97.4 -> 104.4Improved
add global visual feature summary node (ablation)SG2Caps · COCO Karpathy test (ablation, cross-entropy)CIDEr: 104.4 -> 106.6Improved
add global visual feature to baseline without bbox/HOI (ablation)scene graph baseline · COCO Karpathy test (ablation, cross-entropy)CIDEr: 92.6 -> 98.1Improved
add scene graph embeddings to global-feature model (ablation)SG2Caps-G · COCO Karpathy testCIDEr: 99.5 -> 106.6Improved
add HOI graph to pseudolabel inputTSG-trained caption model · COCO Karpathy test (TSG-trained model, different inputs)CIDEr: 31.5 -> 38.6Improved
feed visual scene graph instead of textual scene graph at inferenceTSG-trained caption model · COCO Karpathy test (TSG-trained model, different inputs)CIDEr: 158.6 -> 31.5Hurt
train with CIDEr-based RL after cross-entropySG2Caps · COCO Karpathy test (scene graph only)CIDEr: 104.4 -> 112.3Improved
use SG2Caps instead of SubGC in G-only modescene-graph-only captioner · COCO Karpathy test (scene graph only)CIDEr: 98.9 -> 104.4Improved
use scene graph labels only instead of visual-feature modelSG2Caps vs SGAE · COCO Karpathy testtrainable parameters: 41M -> 21MImproved

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