LaneGCN 部署流水线(DMLaneGCN)
面向自动驾驶 motion forecasting 的 LaneGCN 模型 ONNX / TensorRT 部署流水线,在 TensorRT 10.x 上实现 ≤ 10ms 端到端推理延迟,包含 ONNX 导出、动态 shape 引擎构建、weight-only INT8 量化与可视化验证。
I enjoy making things. Here are a selection of projects that I have worked on over the years.
面向自动驾驶 motion forecasting 的 LaneGCN 模型 ONNX / TensorRT 部署流水线,在 TensorRT 10.x 上实现 ≤ 10ms 端到端推理延迟,包含 ONNX 导出、动态 shape 引擎构建、weight-only INT8 量化与可视化验证。
基于 AIPC 与多模态大模型的端侧 AI 草莓品控工具。结合 YOLO11 目标检测/分割、RAG 知识库检索与多模态大模型,在搭载骁龙 X Elite 的 AIPC 端侧实现草莓成熟度与病虫害的精准识别,可提升优果率 20% 以上。