AI Paper Atlas

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

NNEval: Neural Network based Evaluation Metric for Image Captioning

Naeha Sharif, Lyndon White, Mohammed Bennamoun, Syed Afaq Ali Shah

ECCV 2018 · Vision-language and multimodal › Human and embodied modeling · approach: Neural image captioning · 22 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
replace fixed linear combination of SPICE and CIDEr with a learned neural metricSPIDEr · COMPOSITE (caption-level correlation)Pearson correlation: 0.467 -> 0.532Improved
combine several metric scores with a learned network instead of using SPICE aloneSPICE · COMPOSITE (caption-level correlation)Pearson correlation: 0.475 -> 0.532Improved
learned metric instead of SPIDEr on rank correlationSPIDEr · COMPOSITE (caption-level correlation)Kendall tau: 0.381 -> 0.404Improved
learned metric instead of SPIDEr on SpearmanSPIDEr · COMPOSITE (caption-level correlation)Spearman correlation: 0.495 -> 0.524Improved
learned composite metric instead of METEOR/SPIDErMETEOR · PASCAL-50Saverage accuracy (%): 78.8 -> 80.5Improved
learned composite metric on Human-Human Incorrect pairs vs CIDErSPIDEr · PASCAL-50S (HI pairs)accuracy (%): 98.5 -> 99.0Improved
learned metric vs METEOR on hardest Human-Human Correct pairsMETEOR · PASCAL-50S (HC pairs)accuracy (%): 61.1 -> 60.4Hurt
learned metric vs SPICE on Machine-Machine pairsSPICE · PASCAL-50S (MM pairs)accuracy (%): 71.6 -> 70.4Hurt
learned metric instead of SPIDEr under caption distractionsSPIDEr · Distraction tasks (Hodosh-style forced choice)average accuracy (%): 89.3 -> 91.9Improved
learned metric raises worst-case accuracy vs SPIDErSPIDEr · Distraction tasks (Hodosh-style forced choice)worst-case accuracy (%): 76.6 -> 85.8Improved
learned metric vs CIDEr on Replace Scene distractorsCIDEr · Distraction tasks (Replace Scene)accuracy (%): 95.0 -> 91.8Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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