CIDEr Score Calculator
Calculate the CIDEr score used to evaluate image captioning model outputs.
Inputs
CIDEr Score
7.2000
Scaled Similarity
7.2000
Step by step
Average TF-IDF cosine similarity
across 5 references
= 0.7200
CIDEr score (scaled ×10)
0.7200 × 10
= 7.2000
How it works
CIDEr (Consensus-based Image Description Evaluation) measures image captioning quality by computing TF-IDF weighted n-gram vectors for candidate and reference captions, then calculating their cosine similarity. It rewards captions that use words common in references but rare across the corpus, capturing consensus among annotators.
Formula
CIDEr
CIDEr_n = (1/M) * sum(cos_sim(tfidf_candidate, tfidf_ref_j)) * 10
- M
- Number of reference captions
- cos_sim
- Cosine similarity of TF-IDF vectors
Frequently Asked Questions
Why does CIDEr use TF-IDF weighting?
TF-IDF downweights common words (like 'the', 'a') and upweights informative words (like 'surfboard', 'sunset'), so CIDEr rewards captions that capture the distinctive content of an image rather than just generic descriptions.
Why is CIDEr multiplied by 10?
The scaling factor of 10 is a convention to make scores more readable — raw cosine similarities are typically between 0 and 1, so multiplying by 10 gives a more intuitive range.