Skip to content

Commit efe2d5e

Browse files
xypkentxypkent
committed
Improve README product narrative
1 parent 553a382 commit efe2d5e

2 files changed

Lines changed: 66 additions & 4 deletions

File tree

‎README.en.md‎

Lines changed: 33 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -29,6 +29,38 @@
2929
- **[2026.06.20]** 🧩 Released the **[datamagic-video skill](./datamagic-video/)** — reusable guidance that teaches AI coding agents (Claude Code, Cursor, Codex) to turn tabular data into narrated data videos.
3030
- **[2026.06.18]** 📄 Our paper **"DataMagic: Transforming Tabular Data into Data Insight Video"** has been accepted to **VLDB 2026 Demo Track** and is now available on [arXiv](https://arxiv.org/abs/2606.20388).
3131

32+
## Why DataMagic?
33+
34+
Most teams already have tables. The hard part is turning those tables into something other people can quickly understand: finding what is worth saying, choosing the right charts, arranging the story, writing narration, timing animations, and producing a video people can actually watch.
35+
36+
DataMagic is built to remove that repetitive work. It helps analyze the data, surface useful insights, and turn them into an editable narrated video you can preview, refine, export, and share.
37+
38+
Today's tools usually cover only one part of the job:
39+
40+
- **Static visualization tools** can generate charts, but not a coherent narrated story.
41+
- **BI dashboards** support exploration, but audiences still need someone to explain the narrative.
42+
- **Traditional video editors** can produce polished results, but you still need to prepare charts, scripts, and animations yourself.
43+
- **Pixel-level video generators** such as Seedance, Sora, and Veo can synthesize attractive videos, but they primarily generate pixels rather than data-bound visual elements, making it hard to trust the numbers, labels, and charts.
44+
- **Manual coding with D3, Remotion, or video editors** gives control, but it is slow and hard to scale.
45+
46+
DataMagic is built for this missing middle: **turn raw structured data into an editable, traceable, narrated data video**.
47+
48+
## What Is DataMagic?
49+
50+
DataMagic is an AI-assisted system for authoring data videos from tabular data. Upload a CSV or Excel table, provide an analysis goal or business question, and DataMagic helps analyze the data, surface insights, plan the story, choose charts, draft narration, synchronize animation, preview the result, and export an MP4 video.
51+
52+
The goal is not just to "generate a video." The numbers, labels, and charts in a DataMagic video should remain connected to the original table. Under the hood, **DVSpec** connects visual elements, narration, and animation timing so the result is easier to inspect, edit, and extend.
53+
54+
## What Makes It Different?
55+
56+
| What you want to do | Where common tools fall short | How DataMagic helps |
57+
|---|---|---|
58+
| Turn a table into a video | Analysis, charting, scripting, and editing usually happen in separate tools | One workflow from uploaded data to narrated video |
59+
| Let AI find what matters | Many tools draw charts but do not decide what is worth saying | Analyze the data and organize useful findings into scenes |
60+
| Avoid wrong numbers in AI video | Seedance, Sora, Veo, and other pixel-level models may hallucinate values or labels | Bind chart elements to source data through DVSpec |
61+
| Edit after generation | Regeneration or manual video editing is often required | Preview, edit text, refine with natural language, and keep changes local where possible |
62+
| Share the result quickly | Static charts still need someone to explain them | Export a playable animated data story |
63+
3264
## ⚡ Quick Start
3365

3466
1. **Upload your data** — CSV or Excel table
@@ -41,7 +73,7 @@
4173

4274
**Fast Generation** — Follows the same process as Full Pipeline — AI still handles content planning and narration — but scene rendering uses pre-built visual templates instead of per-scene AI generation, making it significantly faster with less visual customization. Best for users who prioritize speed, such as recurring report production or when visual style is not a primary concern.
4375

44-
**Single Chart** — No full video needed, just one animated chart to illustrate a data insight. Paste your data and quickly generate a single animated chart, ready to embed in a presentation, report, or social media post. Best for users who want to communicate a single data point without a full narrative structure.
76+
**Single Chart** — No full video needed, just one animated chart to discover or explain a focused data point. Paste your data and quickly generate a single animated chart, ready to embed in a presentation, report, or social media post. Best for quick exploration or communicating a local insight without a full narrative structure.
4577

4678
> [!NOTE]
4779
> **Fast Generation** and **Single Chart** are experimental features currently in beta. They work well for typical inputs but may produce unexpected results in edge cases. Feedback and bug reports via Issues are very welcome.
@@ -193,4 +225,3 @@ If you find DataMagic useful in your research or work, please cite:
193225
<div align="center">
194226
<img src="./assets/framework-1.png" width="760" alt="DataMagic system framework">
195227
</div>
196-

‎README.md‎

Lines changed: 33 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -29,6 +29,38 @@
2929
- **[2026.06.20]** 🧩 发布 **[datamagic-video skill](./datamagic-video/)**，可复用的指导，教 AI 编程智能体（Claude Code、Cursor、Codex）把表格数据变成带旁白的动态数据视频。
3030
- **[2026.06.18]** 📄 我们的论文 **"DataMagic: Transforming Tabular Data into Data Insight Video"** 被 **VLDB 2026 Demo Track** 录用，现已上线 [arXiv](https://arxiv.org/abs/2606.20388)。
3131

32+
## 为什么需要 DataMagic？
33+
34+
很多时候，我们手里已经有了一份表格，却还要花大量时间回答这些问题：数据里到底有什么值得讲？应该画什么图？怎么把图串成一个清楚的故事？如果要给团队、客户或课堂展示，还要写旁白、配动画、录屏或剪视频。
35+
36+
DataMagic 想解决的就是这个过程中的重复劳动：让 AI 先帮你分析数据、发现重点，再把这些重点组织成一段可以直接播放、修改和分享的数据视频。
37+
38+
今天的工具通常只覆盖其中一段：
39+
40+
- **静态可视化工具** 可以生成图表，但很难形成带旁白和动画的完整故事。
41+
- **BI 仪表盘** 适合探索，但观众仍然需要有人解释结论和叙事线。
42+
- **传统视频编辑工具** 能做出好看的结果，但需要你先准备图表、脚本和动画。
43+
- **像素级 AI 视频生成模型**（如 Seedance、Sora、Veo 等）可以生成看起来不错的视频，但它们主要生成像素结果，不天然绑定底层数据，因此难以保证数字、标签和图表内容准确。
44+
- **手写 D3、Remotion 或剪辑视频** 控制力强，但成本高、速度慢，不适合日常数据汇报规模化使用。
45+
46+
DataMagic 面向的正是这个缺口：**把原始结构化数据转化为可编辑、可追溯、带旁白的数据视频**。
47+
48+
## DataMagic 是什么？
49+
50+
DataMagic 是一个 AI 辅助的数据视频生成系统。你可以上传 CSV 或 Excel 表格，给出一个分析目标或业务问题，DataMagic 会帮助分析数据、挖掘洞察、规划故事结构、选择图表、生成旁白、同步动画、预览结果，并导出 MP4 视频。
51+
52+
它的核心不是简单地“生成一个视频”，而是让视频里的数字、标签和图表都能对应回原始数据。技术上，DataMagic 使用 **DVSpec** 把视觉元素、旁白和动画时序连接起来，让结果更容易检查、修改和继续生成。
53+
54+
## DataMagic 有什么不同？
55+
56+
| 用户想做的事 | 常见工具的问题 | DataMagic 的做法 |
57+
|---|---|---|
58+
| 从一张表快速变成视频 | 分析、作图、写稿、剪辑通常分散在不同工具里 | 从上传数据到生成带旁白视频的一体化流程 |
59+
| 让 AI 帮忙发现重点 | 很多工具只负责画图，不负责判断哪些数据值得讲 | 自动分析数据并组织成可讲述的场景 |
60+
| 避免 AI 把数字画错 | Seedance、Sora、Veo 等像素级视频模型可能幻觉数值或标签 | 图表元素通过 DVSpec 绑定到源数据 |
61+
| 生成后还能改 | 往往需要重新生成或手动剪辑 | 支持预览、文本编辑、自然语言调整，并尽可能保持局部更新 |
62+
| 直接用于汇报和分享 | 静态图表仍然需要人工讲解 | 导出可播放、可分享的动态数据故事 |
63+
3264
## ⚡ 快速上手
3365

3466
1. **上传数据**：CSV 或 Excel 表格
@@ -41,7 +73,7 @@
4173

4274
**快速生成**：流程与完整流程相同，AI 同样负责规划内容和生成旁白，区别在于场景渲染使用预置视觉模板而非逐场景 AI 生成，因此速度更快，但视觉上没有完整流程那么精细。适合对效率要求更高的用户，比如需要定期产出报告、或对视觉风格没有特别要求的场景。
4375

44-
**单图表生成**：不需要完整视频，只需要一张能说明数据的动态图表。粘贴数据，快速生成一张带动效的图表，直接用在 PPT、对外报告或社交媒体里。适合想快速表达单个数据洞察、不需要完整叙述结构的用户。
76+
**单图表生成**：不需要完整视频，只需要一张能帮助发现或说明重点的动态图表。粘贴数据，快速生成一张带动效的图表，直接用在 PPT、对外报告或社交媒体里。适合快速探索一个数据重点，或表达一个不需要完整叙事结构的局部洞察。
4577

4678
> [!NOTE]
4779
> **快速生成** 与 **单图表生成** 目前处于测试阶段（Beta）。常见用例下运行正常，但在一些边缘情况下可能出现非预期结果。欢迎通过 Issues 反馈问题。
@@ -187,4 +219,3 @@ cp -r DataMagic/datamagic-video ~/.claude/skills/datamagic-video # Claude Code
187219
<div align="center">
188220
<img src="./assets/framework-1.png" width="760" alt="DataMagic 系统架构图">
189221
</div>
190-

0 commit comments

Comments
 (0)