Hello, bean skin fans , Narrative visualization is a hot research direction in academic circles in recent years , Today, xuanhun brings you to share 《 Narrative visualization and data video 》
author ： Mysterious soul
Narrative visualization is a hot research direction in academic circles in recent years , The content of this paper is a review of some papers （ See the references at the end of the text for details ） Extraction and sorting of , It's about throwing a brick at a jade . Finally, the paper puts forward the development idea of a data video generation tool for ordinary users , Welcome to discuss .
1. Narrative Visualization （Narrative Visualization）
Narrative visualization is simply to tell a story in a visual form . The common form is information graph , Data cartoon , Data, video, etc .
1.1 Visual narrative grammar （Visual Narrative Grammar）
Visual narrative grammar （VNG） A sequential image understanding model integrating linguistic theory and psychological experiment is proposed . VNG Designed to describe the narrative structure of continuous images .
There are four components of narrative structure ：Establisher(E), Initial(I), Peak(P), Release(R).
- Establisher (E): A sequence of actions or events that provide reference information without participating in the narrative
- Initial (I): “ Initiate an action or event ” Sequence
- Peak (P): The climax of an event
- Release (R): The sequence after orgasm
VNG In fact, it is developed from the traditional drama creation theory , It is similar to the writing method we usually talk about .
Pictured above , This is a simple cartoon ,Establisher Stage builds the scene , He explained the time and place of the character ; The second picture begins with an event —— The protagonist couldn't sleep and began to think about things .Peak Stage , The character began to wonder whether to get out of bed and take a shower , It's an iterative process . Last , Dramatis personae Release Phase Gong still decided to sleep .
In order to show a more complex narrative structure , The four basic components are usually built in a tree nested structure . It's kind of like a novel , Small events will be split under the overall framework , Only in this way can there be ups and downs , The effect of repeated orgasms . Let's analyze several narrative patterns .
1.2 Typical narrative mode
[Phase X (Establisher) – (Initial) – Peak – (Release)]
A typical narrative pattern , Whether the overall narrative structure or local sub processes , Only Peak Is a must , Other ingredients are optional .
Like the cartoon above , The whole is divided into three parts ：
Initial： Describe two boxers competing in the ring
Peak： Climax part , A beautiful blow from the boxer on the left , Knock down your opponent
Release： The ending
This is a tree nested structure , Each stage can extend its own sub process . Finally, the overall narrative structure is ：I-P-E-I-Ref-P-R
Conjunction schema [Phase X X1 - X2 -... Xn]
Throughout the narrative , We often need to use different ways to present local information , Decompose the content to be expressed as a whole , Can better highlight the key points 、 Express intention .
If several consecutive images play the same grammatical role in the components of this stage , We can present the final overall concept through different combinations . Here's the picture , These continuous images on the left create the conceptual equivalent of the right panel .
The concentrated combination in the figure above is what we can often see in comics , Grammatically, it is described as follows ：
- （a） An action or event （a- Conjunction ）,
- （b） The characters in the scene （E- Conjunction ）,
- （c） Part of a single role （N- Conjunction ）,
- （d） Different semantic related elements （S- Conjunction ）
2. Data video
Data video is a common visual narrative form , By customizing the animation of graphics and charts and combining auditory sound effects , So as to tell the story in the data . Data video is a very attractive way of data narration , It can provide diverse information to the audience in a short time .
For example, below is a video , Through the cooperation of China, Japan and South Korea GDP Data changes , It shows the economic development trend since China's reform and opening up , At the same time, it is accompanied by patriotic music to highlight the patriotic theme .
In order to better understand the content and structure of data video , Relevant papers are collected 50 A typical data video is analyzed qualitatively .
2.1 Narrative structure of data video
It is observed that the data video is also layered narrative , And can be further decomposed into sub units . The following figure shows the statistical results , Use regular expressions to represent narrative patterns .
Access link ：https://www.shutterstock.com/...
You can watch this long 2.07 Minutes of data video . The video follows （EI+PR+） Narrative structure pattern , Comprehensive application of voice narration 、 Video clip 、 Data visualization and attention cues , It tells us a powerful story about the evolution of the relationship between human beings and the media .
Establisher Part shows the theme , Show that the video is about “ The power of video ”, Make a rapid transition to Initial Stage , Include 4 Basic units . The original 3 A unit describes a basic fact , Statistics on the rise in video usage . At the beginning of the fourth unit , The narrator asks “ Why video ？“ Then there are fragments of daily life , By reflecting on the information that has been presented , To answer a question in one sentence Peak Create tension at this stage .Peak Stage uses exciting music and animation effects to attract the attention of the audience . then , The video continues to present more facts , Support Peak The answer given in . This is through 4 A publishing unit completes , All these units include data visualization . Dynamic charts 、 Highlighting a single bar in the bar chart and the continuous change of the year range are common attention tips in data visualization .
The most common narrative pattern is E+I+PR+ Pattern ,34% Your video uses this mode .
These structures contain a climax unit （P）, There are no ending units . The data video of this narrative structure presents some facts in the data , Leave an audience “ problem ” or “ Things that need to be thought about ”, To complete the narrative .
“EI+” and “ER+”
Flat data presentation , Often rely on the charm of the data itself to attract the audience .
2.2 Composition analysis
If you count the time distribution of these fragments , We will find that Initial Part of the time is the longest , On average, it accounts for more than 60% Time for , And such Initial The part of is often short in traditional video , This shows that the structure of data video is different from that of traditional video .
The researchers also made statistics on the proportion of data visualization and attention prompt information in each structural component （ The figure below ）, Found in Initial In video clip ,60% The content of contains the elements of data visualization , This phenomenon shows that visualization is often used to pave the way for events , Not necessarily only for the presentation of event results . Besides , The distribution of attention cue information is basically similar to that of data visualization , It shows that they are directly related , Data visualization often needs these attention cues to enhance their expressiveness . About 20% Your video will start with a question ,30% The video will provide some additional instructions or guidance information at the end .
2.3 Use of visual elements of data video
In data video , Visual presentation of data is the main means of storytelling . Although the average duration of data video is 3 minute （1 To 7.5 minute ）, But the average number of data visualizations presented is 6 Time （1 To 19 Time ）. On average, , Total duration of data video 48% For data visualization .72% The average video data depends only on 5 There are different types of visualization （ chart 2- Left ）. More than half of the data video duration includes data visualization , But they are limited to 3 In various forms （ Scatter plot 、 Bar charts and maps ）.
We identified nine main types of visual and auditory effects , Designed to attract the attention of the audience .
The first three most commonly used effects are animation 、 appear / Disappear and highlight .
3. How to make data video
In terms of tools and Ecology , At present, there is no independent tool for the production of data video , It is not subdivided from the big branch of video creation .
The methods commonly used at present
- Chart generation tool + Video editing tools
Some chart libraries provide simple dynamic chart making tools , However, it can not effectively meet the needs of visual narration . For relevant recommendations, please refer to Zhihu ：https://www.zhihu.com/questio...
At present, the types and forms of chart tools out of the box are very limited , This is also what we see. Most users' data videos are column charts 、 One of the reasons for the line diagram .
- Image editing tools + Video production tools
Almost all well-made data videos are drawn by professional designers , Combined with video editing .
In some papers , You can see the research in this direction , such as DataClips（ https://fereshtehamini.github...
）, But the tools that are actually publicly available do not , This market is still a blank .
## 4. oriented C End crowd （ Content creators ） The idea of a data video production and publishing tool
（ The following software interface is only an auxiliary example , Not a screenshot of the actual data video production tool ）
We can see the frequency of creation and release of video content , In quite a large crowd , Has gone beyond the creation of text content . The theme of data video is prominent , Various forms of presentation , The narrative structure is simple and lively , It has natural creative advantages in the field of short video publishing .
It is essentially a story design tool
Video is just an end product （ Or one of the products ）, The core function is how to design and arrange a story . This and PPT It's conceptually the same , We use to design a story how to tell 、 How to present 、 How to interact , final “ Export video ” Just a feature .
Data visualization is the core presentation form
This tool will be the traditional visual presentation tool 、 The core functions of data analysis tools are perfectly combined with story creation . Visualization will play the following roles here ：
- There are various forms of presentation . Statistical charts can be 、 chart 、Gis And other various forms of display , Get rid of class PPT Tools focus on text 、 picture 、 The presentation form of graphics .
- A new narrative form . With the help of visual presentation form and lens language , We can have a new multi-dimensional narrative . For example, the full canvas narrative below , To break the PPT The point to line display and narrative mode of class tools , Expanded “ Noodles ”.
Of course, we can go further , Use 3d The way to narrate .
Intelligence is the source power to continuously meet the needs of users
Only smart enough , And evolving tools can continue to meet the needs of users . Intellectualization is embodied in the following aspects ：
- Intelligent data analysis capability
- Intelligent text analysis ability
- Visual automatic generation capability
- Automatic story arrangement ability
- Automatic material recommendation ability
Research on the application of narrative visualization in video content scene is of great significance not only to improve the narrative ability of data visualization itself, but also to improve the expressiveness of video content . The semantic expression of data visualization and video can learn from each other . In this direction , Both content output and academic research are at an initial stage , I hope this article can give you some inspiration .