Repository navigation
feat(OmniH2O): 重写详情笔记，新增十七幕讲解动画、三个交互演示与配音视频 - #636
Merged
Merged
Conversation
- 笔记对照 arXiv v1 的 LaTeX 源码、PMLR 定稿与官方仓库 LeCAR-Lab/human2humanoid 逐条核查重写，记下代码与论文的出入、表 3 与附录表 17 对不上的两处 - 十七幕动画：H1 用官方样例 stable_punch.pkl 的左手出拳离线渲染成 3D 贴图（蹲姿版与原始重定向是我们照论文描述做的，已标明）；用上论文图 1、3–11 与官方 README 硬件图 - 三个交互演示：三个点浏览器、抬脚奖励与惩罚课程、表 1–3 - 配音视频 13 分 25 秒（四段并行渲染 24,143 帧），网页版 -crf 37 压到 14.5 MB，海报取自封面 - 路线图 OmniH2O 节点加 🎬，重新生成 _data/papers.json；测试登记十七幕并核对动画、演示、旁白与具体实例共用的数字 - AGENTS.md、视频 README、发布清单记下 omnih2o 的接线与压缩参数 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019XgFzxdYKHJTJgs5FJfMyB
- 片头「还能拿遥操作数据」被 TTS 读成「拿遥，操作数据」（遥后 240 ms 无声），改句；全篇十处「遥操作」用基频跟踪量过，都在 50–80 ms - 真机画面换成 omni.human2humanoid.com 项目页视频剪的 10 段片段（写毛笔字、四种 VR 遥操作、上下身任务、MDM、抗干扰、户外、GPT-4o 两段、扩散策略四个任务、三个失败案例，共 3.1 MB），去掉不再用的论文图 1、4、6、7 - kit.js 新增 svgClip / clipSince：网页上播到那一幕才下载、离开就停；离线视频 stage.html 按时间线算片段时刻，render.mjs 转 VP9 逐帧 seek，另加 chunk 分段模式 - check_layout.mjs 加 story near（间距不到 3px 也报），修第 3–8、10、16 幕贴边贴字（第 5 幕的 120、40 贴着右栏） - AGENTS.md / 视频 README 写下「真机效果用项目页的视频片段」与读音停顿的查法；整片待确认后重渲 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019XgFzxdYKHJTJgs5FJfMyB
- TTS 的切词把「跟头」当成一个词，念成「只，跟头」；改句后切成「跟踪 | 头」，重合成这一句，「遥操作」间隙仍是 50 ms - 第 14 幕右栏的同一句标注一并改成「只跟踪」 - 视频 README 记下：全篇切词按关键字逐个看上下文；H2O 已发布的片尾预告也是这句，未改 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019XgFzxdYKHJTJgs5FJfMyB
- render.mjs chunk 四段并行重渲 24,136 帧（13 分 25 秒，音视频 804.53 / 804.53 s），含片头两处读音修正、排版修正与第 10、14–17 幕的项目页真机片段 - 网页版 -crf 37 压到 15.1 MB（初版 14.5 MB），替换笔记 media 里的视频，更新下载按钮与兜底链接的体积 - 发布清单 027 OmniH2O 打勾，下一集为 028 MaskedMimic；AGENTS.md、视频 README 记下压缩参数与渲染耗时 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_019XgFzxdYKHJTJgs5FJfMyB
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_019XgFzxdYKHJTJgs5FJfMyB