List your company

Image Captioning on UCM

0.918BLEU-1Ours
0.822320.847160.8720.89684Dec 29, 2025

Evaluation results

MethodLinks
OursVisual Encoder=CLIP-Cap-4, Text Decoder=LlamaV22025.120.9180.8680.8290.7980.4890.8323.7010.547
SkyEyeGPTVisual Encoder=EVA-G, Text Decoder=LlamaV2-Chat2025.120.9070.8570.8160.7840.4620.7952.368—
RSGPTVisual Encoder=EVA-G, Text Decoder=Vicuna2025.120.8610.7910.7230.6570.4220.7833.332—
RS-CapRetVisual Encoder=CLIP-Cap-4, Text Decoder=LlamaV2, Finetuned=true2025.120.8430.7790.7220.670.4720.8173.5480.525
RS-CapRetVisual Encoder=CLIP-Cap-4, Text Decoder=LlamaV2, Finetuned=false2025.120.8330.760.6990.6450.4470.7863.4290.525
MLCA-NETVisual Encoder=VGG16, Text Decoder=LSTM2025.120.8260.770.7170.6680.4350.7723.240.473