<![CDATA[知来者逆的博客]]>

<![CDATA[知来者逆的博客]]>https://blog.youkuaiyun.com/matt45mzh-cnhttps://blog.youkuaiyun.com/<![CDATA[MiniCPM-o 4.5：实时能听能看能说的多模态模型]]>https://blog.youkuaiyun.com/matt45m/article/details/166896600https://blog.youkuaiyun.com/matt45m/article/details/166896600matt45mWed, 30 Sep 2026 16:01:45 +0800<![CDATA[计算机视觉——从环境配置到跨线计数的完整实现基于 YOLOv12 与质心追踪器的实时人员监控系统]]>https://blog.youkuaiyun.com/matt45m/article/details/153833981https://blog.youkuaiyun.com/matt45m/article/details/153833981matt45mFri, 24 Oct 2025 17:04:35 +0800<![CDATA[计算机视觉——从YOLO系列演进到YOLOv12架构创新、注意力机制优化、推理实践与性能基准]]>https://blog.youkuaiyun.com/matt45m/article/details/153117152https://blog.youkuaiyun.com/matt45m/article/details/153117152matt45mSun, 12 Oct 2025 14:19:09 +0800<![CDATA[YOLOv8提升小目标检测策略——专为微小目标检测设计的 IoU 替代方案NWD原理解析、YOLO代码集成与验证]]>https://blog.youkuaiyun.com/matt45m/article/details/151826375https://blog.youkuaiyun.com/matt45m/article/details/151826375matt45mThu, 18 Sep 2025 13:23:41 +0800<![CDATA[YOLOv8提升小目标检测策略 —— EFC（增强层间特征关联）与 FCM（特征互补映射）的原理与算法代码实现]]>https://blog.youkuaiyun.com/matt45m/article/details/151753466https://blog.youkuaiyun.com/matt45m/article/details/151753466matt45mTue, 16 Sep 2025 12:33:36 +0800<![CDATA[YOLOv8提升小目标检测策略——详细记录从 HCF-Net 的 DASI/PPA 到 SPD-Conv、MSCA 及 BoTNet加EMA优化原理到代码实现方法]]>https://blog.youkuaiyun.com/matt45m/article/details/151612634https://blog.youkuaiyun.com/matt45m/article/details/151612634matt45mFri, 12 Sep 2025 14:05:26 +0800<![CDATA[视觉语言模型应用开发——Qwen 2.5 VL模型视频理解与定位能力深度解析及实践指南]]>https://blog.youkuaiyun.com/matt45m/article/details/151319336https://blog.youkuaiyun.com/matt45m/article/details/151319336matt45mMon, 08 Sep 2025 14:52:48 +0800<![CDATA[视觉语言模型应用开发——Qwen 2.5 视觉语言模型的零样本学习能力在多模态内容审核中的实践研究]]>https://blog.youkuaiyun.com/matt45m/article/details/150925259https://blog.youkuaiyun.com/matt45m/article/details/150925259matt45mWed, 27 Aug 2025 16:04:52 +0800<![CDATA[多模态大模型——Qwen2.5-Omni端到端实时多模态智能的技术架构与实现机制]]>https://blog.youkuaiyun.com/matt45m/article/details/150852109https://blog.youkuaiyun.com/matt45m/article/details/150852109matt45mTue, 26 Aug 2025 13:39:35 +0800<![CDATA[大语言模型应用开发——利用OpenAI函数与LangChain结合从文本构建知识图谱搭建RAG应用全流程]]>https://blog.youkuaiyun.com/matt45m/article/details/149810818https://blog.youkuaiyun.com/matt45m/article/details/149810818matt45mSat, 23 Aug 2025 20:20:10 +0800<![CDATA[视觉语言大模型应用开发——基于 CLIP、Gemini 与 Qwen2.5-VL 的视频理解内容审核全流程实现]]>https://blog.youkuaiyun.com/matt45m/article/details/150611497https://blog.youkuaiyun.com/matt45m/article/details/150611497matt45mFri, 22 Aug 2025 15:47:55 +0800<![CDATA[面向复杂场景的人脸识别——AdaFace 模型的理论基础与从数据到部署的实践路径]]>https://blog.youkuaiyun.com/matt45m/article/details/150580170https://blog.youkuaiyun.com/matt45m/article/details/150580170matt45mThu, 21 Aug 2025 12:25:48 +0800<![CDATA[YOLOv8 姿态估计模型的 OpenVINO 部署——从 IR 格式转换、精度评估到 NNCF 量化部署及性能对比]]>https://blog.youkuaiyun.com/matt45m/article/details/150550427https://blog.youkuaiyun.com/matt45m/article/details/150550427matt45mWed, 20 Aug 2025 11:45:50 +0800<![CDATA[姿态估计进阶：从 YOLO-NAS Pose 到 YOLOv8 Pose的技术探索与实践]]>https://blog.youkuaiyun.com/matt45m/article/details/150529570https://blog.youkuaiyun.com/matt45m/article/details/150529570matt45mTue, 19 Aug 2025 16:59:10 +0800<![CDATA[从 3DCNN 到 Vision Transformer——视频动作识别技术的演进与 ViT 的突破性应用]]>https://blog.youkuaiyun.com/matt45m/article/details/150487808https://blog.youkuaiyun.com/matt45m/article/details/150487808matt45mMon, 18 Aug 2025 11:56:42 +0800<![CDATA[微调 AnomalyCLIP——基于对象无关提示学习与全局 - 局部优化的零样本异常检测框架性能验证]]>https://blog.youkuaiyun.com/matt45m/article/details/149930803https://blog.youkuaiyun.com/matt45m/article/details/149930803matt45mSun, 17 Aug 2025 14:54:32 +0800<![CDATA[基于CNN 的人体动作检测：训练过程、结果可视化及模型测试]]>https://blog.youkuaiyun.com/matt45m/article/details/150425777https://blog.youkuaiyun.com/matt45m/article/details/150425777matt45mFri, 15 Aug 2025 16:23:09 +0800<![CDATA[视觉语言大模型应用开发——基于Qwen2.5-VL 实现视觉语言模型在目标检测中的层级结构与实现方法]]>https://blog.youkuaiyun.com/matt45m/article/details/150392157https://blog.youkuaiyun.com/matt45m/article/details/150392157matt45mThu, 14 Aug 2025 15:51:01 +0800<![CDATA[视觉语言模型的空间推理缺陷——AI 在医学扫描中难以区分左右]]>https://blog.youkuaiyun.com/matt45m/article/details/150014905https://blog.youkuaiyun.com/matt45m/article/details/150014905matt45mThu, 07 Aug 2025 15:45:02 +0800<![CDATA[LLM开发——语言模型会根据你的提问方式来改变答案]]>https://blog.youkuaiyun.com/matt45m/article/details/149943044https://blog.youkuaiyun.com/matt45m/article/details/149943044matt45mTue, 05 Aug 2025 22:32:16 +0800