ArticleMetadataMain page

source · 2026-04-25

MMPose / RTMPose（GitHub repository）：MMPose 提供 2D/3D 姿态估计工具链，RTMPose 强调实时部署。

#near-cvpr-2025#video-understanding#secondary-source

MMPose / RTMPose（GitHub repository）：MMPose 提供 2D/3D 姿态估计工具链，RTMPose 强调实时部署。

资料来源：GitHub repository
发表 / 发布日期：2023-01-01
资料加入日期：2026-04-25

一句话结论

MMPose 提供 2D/3D 姿态估计工具链，RTMPose 强调实时部署。

资料定位

这是从体育机器视觉文献清单中拆出的独立来源。当前 first-pass 编译先把它放进 raw/source 证据层，后续可按重要性升级为深分析笔记。

对体育 AI 子线的价值

它适合作为体育姿态、训练反馈和移动端原型的工程基线。

局限或疑问

  • 当前是 bibliography-driven first-pass source note，优先保证来源可追溯和方向定位。
  • 重要 paper 后续应补 analysis.md、关键图页、实验设置和指标细节。

原始材料

  • raw/ingest/2026-04-25-mmpose/source.md
  • raw/ingest/2026-04-25-mmpose/paper-text.md
  • raw/ingest/2026-04-25-mmpose/links.yaml

相关页面

Metadata

{
  "id": "2026-04-25-mmpose",
  "type": "source",
  "title": "MMPose / RTMPose（GitHub repository）：MMPose 提供 2D/3D 姿态估计工具链，RTMPose 强调实时部署。",
  "status": "reviewed",
  "created": "2026-04-25",
  "updated": "2026-04-25",
  "venue": "GitHub repository",
  "published_at": "2023-01-01",
  "ingested_at": "2026-04-25",
  "tags": [
    "near-cvpr-2025",
    "video-understanding",
    "secondary-source"
  ],
  "note_status": "reviewed",
  "source_type": "repository",
  "authors": [],
  "canonical_links": [
    "https://github.com/open-mmlab/mmpose"
  ],
  "raw_entry": "raw/ingest/2026-04-25-mmpose/",
  "topics": [
    "topics/sports-ai-video-understanding",
    "topics/sports-ai-roadmap",
    "topics/video-understanding"
  ],
  "entities": [
    "entities/sportsmot"
  ],
  "claims": [],
  "questions": []
}