IGINet

Reported on 7 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing7 results

Image CaptioningonCOCO (Common Objects in Context)
BLEU-1
81.1
Image CaptioningonCOCO (Common Objects in Context)
BLEU-2
65.9
Image CaptioningonCOCO (Common Objects in Context)
BLEU-3
51.7
Image CaptioningonCOCO (Common Objects in Context)
BLEU-4
39.9
Image CaptioningonCOCO (Common Objects in Context)
CIDEr
131
best: 143.7 (ExpansionNet v2)
Image CaptioningonCOCO (Common Objects in Context)
METEOR
29.4
Image CaptioningonCOCO (Common Objects in Context)
ROUGE-L
59.2