项目精选
-disaster-vqa-response-system
基于视觉语言模型的灾害图像问答系统
01
项目亮点
- 实时推理与置信度
- 支持大型 VLM 扩展
- FastAPI 后端
- AI 辅助前端仪表盘
FACTS
来自当前修订的事实
- repository owner
- gee-46
- repository name
- -disaster-vqa-response-system
- description
- "An adaptable, AI-powered Disaster VQA system that performs real-time visual assessment of disaster imagery. By defaulting to the highly efficient BLIP model for rapid inference—while maintaining full architectural support for massive Vision-Language Models like LLaVA/Qwen-VL—it delivers contextual answers and confidence scores to accelerate emerge
- primary language
- CSS
- license
- Apache-2.0
- topics
- deep-learning, machine-learning, machine-learning-algorithms, ollama, opencv, python, qwen2-5, vibe-coding
- source count
- 1