车载、路边、无人机 - 标签 - Zhaoylee's Blogs https://zhaoylee.github.io/Blogs_lovelt/tags/%E8%BD%A6%E8%BD%BD%E8%B7%AF%E8%BE%B9%E6%97%A0%E4%BA%BA%E6%9C%BA/ 车载、路边、无人机 - 标签 - Zhaoylee's Blogs Hugo -- gohugo.iozh-CNThu, 12 Mar 2026 10:25:13 +0800 MonoCT: Overcoming Monocular 3D Detection Domain Shift with Consistent Teacher Models https://zhaoylee.github.io/Blogs_lovelt/posts/monoct-overcoming-monocular-3d-detection-domain-shift-with-consistent-teacher-models/ Thu, 12 Mar 2026 10:25:13 +0800 zhaoylee https://zhaoylee.github.io/Blogs_lovelt/posts/monoct-overcoming-monocular-3d-detection-domain-shift-with-consistent-teacher-models/

🏛️ 会议/期刊：ICRA
📅 发表年份：2025
💻 开源代码：GitHub 链接
📄 论文题目：MonoCT: Overcoming Monocular 3D Detection Domain Shift with Consistent Teacher Models

0. 一句话总结 (TL;DR)

(这篇论文用什么方法，解决了什么问题，达到了什么效果)
MonoCT 提出了一种基于一致性教师模型（Consistent Teacher）的半监督自适应框架，通过在目标域（Target Domain）引入伪标签一致性约束，有效解决了单目 3D 检测在不同数据集间迁移时的深度估计偏差问题。

]]>