English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Current metrics for video captioning are mostly based on the text-levelcomparison between reference and candidate captions. However, they have someinsuperable drawbacks, e.g., they cannot handle videos without references, andthey may result in biased evaluation due to the one-to-many nature ofvideo-to-text and the neglect of visual relevance. From the human evaluator'sviewpoint, a high-quality caption should be consistent with the provided video,but not necessarily be similar to the reference in literal or semantics.Inspired by human evaluation, we propose EMScore (Embedding Matching-basedscore), a novel reference-free metric for video captioning, which directlymeasures similarity between video and candidate captions. Benefit from therecent development of large-scale pre-training models, we exploit a wellpre-trained vision-language model to extract visual and linguistic embeddingsfor computing EMScore. Specifically, EMScore combines matching scores of bothcoarse-grained (video and caption) and fine-grained (frames and words) levels,which takes the overall understanding and detailed characteristics of the videointo account. Furthermore, considering the potential information gain, EMScorecan be flexibly extended to the conditions where human-labeled references areavailable. Last but not least, we collect VATEX-EVAL and ActivityNet-FOIldatasets to systematically evaluate the existing metrics. VATEX-EVALexperiments demonstrate that EMScore has higher human correlation and lowerreference dependency. ActivityNet-FOIL experiment verifies that EMScore caneffectively identify "hallucinating" captions. The datasets will be released tofacilitate the development of video captioning metrics. The code is availableat: https://github.com/ShiYaya/emscore.

EMScore: Evaluating Video Captioning via Coarse-Grained and Fine-Grained Embedding Matching