Skip to content
View t0ugh-sys's full-sized avatar
💻
coding
💻
coding
  • China

Block or report t0ugh-sys

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
t0ugh-sys/README.md

t0ugh

视觉算法工程师 | Computer Vision Engineer

专注目标检测、分割、跟踪、OCR 与边缘部署

面向工业视觉与边缘 AI 场景,关注推理优化、跨平台部署与可落地的视频分析链路

Python PyTorch OpenCV YOLO ONNX TensorRT RKNN

Focus:

  • 视觉算法研发与工程落地(Detection, Segmentation, Tracking, OCR)
  • 模型训练、推理优化与部署(PyTorch, YOLO, ONNX, TensorRT, RKNN)
  • 视觉应用服务化与多模态检索(FastAPI, Milvus, LangChain)

Featured Projects:

  • vision-inference-pipeline — 面向 Rockchip RK3588/RK3568 与 NVIDIA GPU 的视频推理部署骨架,覆盖硬解码、预处理、后端适配与 RKNN / TensorRT 部署。
  • vision-inference-benchmark — 面向视觉推理部署的 benchmark 工程,支持基准测试、结果汇总与跨后端对比分析。
  • camera-calibration-lab — 相机标定工具链,支持单目标定、双目标定、角点可视化与去畸变结果验证。
  • DeepRAG — 企业级 RAG 知识库系统,支持混合检索、重排、多租户隔离与服务化部署。

Architecture Preview:

flowchart LR
    A["FFmpeg Demux"] --> B["Decoder<br/>MPP / NVDEC"]
    B --> C["Preprocess<br/>RGA / CUDA"]
    C --> D["Inference<br/>RKNN / TensorRT"]
    D --> E["Postprocess<br/>YOLO Decode / NMS"]
    E --> F["Industrial Vision Applications"]
Loading
GitHub contribution snake

Pinned Loading

  1. BerriAI/litellm BerriAI/litellm Public

    Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthr…

    Python 48.9k 8.5k

  2. airockchip/rknn_model_zoo airockchip/rknn_model_zoo Public

    C 2.5k 428

  3. Anvil Anvil Public

    A tool-use runtime for coding agents, with session management, permissions, and extensible CLI workflows.

    Python

  4. vision-inference-pipeline vision-inference-pipeline Public

    面向 Rockchip RK3588/RK3568 与 NVIDIA GPU 的视频推理部署骨架,覆盖硬解码、预处理、后端适配与 RKNN / TensorRT 部署。

    C++

  5. DeepRAG DeepRAG Public

    基于 FastAPI + Vue 3 的前后端分离 RAG(检索增强生成)系统,支持 Milvus/FAISS 向量存储、BM25 混合检索、Reranker 重排,以及多租户隔离。

    Python 2