Share your thoughts, 1 month free Claude Pro on usSee more

Fine-grained Image Captioning on DetailCaps (test)

72.2CAPTURE

Qwen2.5-VL+TDSR

Updated 3mo ago

Evaluation Results

Method	Links
Qwen2.5-VL+TDSR 2025.10		72.2	72.3	65.2	64.7
LLaVA-1.5+TDSR 2025.10		66.7	66.2	62.4	63.4
Qwen2.5-VL 2025.10		64.7	66.7	62.5	62.3
Mini-Gemini+Ours 2025.04		64.49	68.15	63.92	56.75
LLaVA1.6+Ours 2025.04		63.51	66.95	63.15	55.8
FINECAPTION 2025.10		63.4	63.7	58.1	58.3
VistaLLM 2025.10		63.2	63.5	60.3	59.2
Ferret 2025.10		62.8	63.2	56.4	57.3
Mini-Gemini+IT 2025.04		62.53	66.85	61.4	54.54
Mini-Gemini+Syn 2025.04		62.53	66.68	61.49	54.77
Mini-Gemini 2025.04		62.37	66.77	61.15	54.4
LLaVA1.6+IT 2025.04		61.21	64.76	60.68	53.68
MiniGPT-v2 2025.10		61.2	62.7	55.9	56.8
LLaVA1.6+Syn 2025.04		61.16	65.17	60.24	53.44
LLaVA1.6 2025.04		61.03	64.59	60.39	53.73
Shikra 2025.10		60.5	61.9	55.4	56.4
Alpha-CLIP 2025.10		59.2	61.9	57.2	56.5
KOSMOS-2 2025.10		58.5	60.8	53.1	54.2
LLaVA-1.5 2025.10		58.05	62.2	56.1	52.5
LLaVA1.5+Ours 2025.04		58.05	62.23	56.09	52.51
VisionLLM-H 2025.10		57.9	59.3	53.4	53.9
LLaVA1.5+Syn 2025.04		57.51	62.06	55.54	51.07
LLaVA1.6+PoCa 2025.04		55.38	58.61	54.47	49.61
Mini-Gemini+PoCa 2025.04		55.08	59.73	52.68	49.44
LLaVA-1.5 2025.10		51.98	56.3	48.2	50.4
LLaVA1.5+IT 2025.04		51.98	56.37	48.22	50.44
LLaVA-1.5 2025.10		49.99	55.7	44.4	49.4
LLaVA1.5 2025.04		49.99	55.76	44.43	49.48
LLaVA1.5+PoCa 2025.04		47.38	54.57	40.64	46.28