用二维特征当文本令牌，让视觉语言模型更轻更快用于自动驾驶
MiniDrive: More Efficient Vision-Language Models with Multi-Level 2D Features as Text Tokens for Autonomous Driving
- 将2D视觉特征转为文本令牌输入语言模型
- 仅8300万参数却达顶尖性能，推理效率显著提升
- 适合资源受限的实时自动驾驶系统部署
视觉语言模型（VLMs）在自动驾驶中作为通用端到端模型，通过问答交互完成预测、规划和感知等任务。然而，现有方法多依赖计算开销大的视觉编码器和大型语言模型（LLMs），难以部署于真实场景与实时应用；且多数VLMs无法处理多图像，难以适应自动驾驶中的多摄像头感知。为此，我们提出MiniDrive框架，包含特征工程混合专家（FE-MoE）模块与动态指令适配器（DI-Adapter）。FE-MoE将2D特征高效映射为视觉令牌嵌入，输入语言模型；DI-Adapter使视觉令牌嵌入随指令文本嵌入动态调整，解决以往静态视觉令牌的问题。相比先前工作，MiniDrive在参数量、浮点运算量和响应效率上均达到最先进水平，最小版本仅含8300万参数。
原文摘要 · Abstract (English)
Vision-language models (VLMs) serve as general-purpose end-to-end models in autonomous driving, performing subtasks such as prediction, planning, and perception through question-and-answer interactions. However, most existing methods rely on computationally expensive visual encoders and large language models (LLMs), making them difficult to deploy in real-world scenarios and real-time applications. Meanwhile, most existing VLMs lack the ability to process multiple images, making it difficult to adapt to multi-camera perception in autonomous driving. To address these issues, we propose a novel framework called MiniDrive, which incorporates our proposed Feature Engineering Mixture of Experts (FE-MoE) module and Dynamic Instruction Adapter (DI-Adapter). The FE-MoE effectively maps 2D features into visual token embeddings before being input into the language model. The DI-Adapter enables the visual token embeddings to dynamically change with the instruction text embeddings, resolving the issue of static visual token embeddings for the same image in previous approaches. Compared to previous works, MiniDrive achieves state-of-the-art performance in terms of parameter size, floating point operations, and response efficiency, with the smallest version containing only 83M parameters.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。