December 16th.AliLaunch of a new generation of multi-person 2.6 series of models, which were completely upgraded to professional video production and image creation scenes, with a new one 2.6 being the first national supportrole-playing (as a game of chess)FunctionVideo ModelI don't know. The model also supports the functions of synchronization, multiple lens generation and sound drive, and is the most functional video generation model in the world. 2.6 Al-Aiyun Ying, a network of officials。
1AI HAS THE FOLLOWING OFFICIAL DESCRIPTION:
2.6 Video Generation Upgrade Point
role-playing (as a game of chess)
Everything can be played, and 2.6 can refer to the appearance and sound of the role in the input video, and achieve role-playing, and generate single, multi-person, human-to-material video by the hint。
Multi-photo narrative
2.6 Users ' simple hints can be converted to multiscopy scripts, generating consistent narrative videos with multiple lenses and maintaining the consistency of key messages such as the subject, scene and so on。
Sync Natural Sound
The scenes of multi-person dialogue are stable, real and natural human voice is expressed, sound quality is raised and music and songs are better heard。
15s long video generation
generates a maximum of 15s (reft 10s) to increase the spatial and temporal content of the image to support a more complete narrative。
Audio Drive Generation
2.6 Videos can be generated and multi-photogramd based on the text and audio entered by the user。
2.6. Upgrade the Vincent Chart
High aesthetic and controllability of art style
- Accurate capture of the artistic soul: a deep understanding of the various style key words, a higher reduction, a unified and complete picture。
- Style integration is natural flow: it supports the organic integration of various art styles, the transition smoothing and the creation of unique visual works。
- The details are vividly rich: they are better presented in terms of "mythics, colours, brushes, etc." and give image stories a sense of expression。
The real person's touch, the light is in place
IN THE CREATION OF A REALITY IMAGE, THE PHYSICAL AND PHYSICAL NATURE OF THE PERSON EFFECTIVELY WEAKENS THE “AI SENSE”. THE MODEL COMBINES THE OPTIMIZATION OF THE IMAGE AND THE LIGHT AND PRESENTS A MORE REPRESENTATIVE HUMAN PERSPECTIVE。
Text generation, graphics, illustrations and poster design
A high degree of consistency between content and vision can be achieved through the accurate generation of posters, illustrations, information diagrams, etc., based on long Chinese and English textual content, in conjunction with aesthetic design and alignment。
Image Generation Upgrade Points
Text Mixed Output
Multiple images and text are produced in an integrated manner, with the ability to reason and achieve a structured and logical visual narrative。
Multigraph Integration Generation
Multiple images are randomly referenced, grouped or replaced, combining multiple visions and generating new and creative images。
Coherence at the commercial level
Maintaining a high degree of consistency in roles, styles or elements and expanding the use of scenarios at the business level。
Aesthetic factor migration
Draws creative ideas such as colour, style, structure, etc. of reference maps to enable aesthetic reference generation。
The camera and the light control
The lens view, near vision and light detail can be specified to achieve precise regulation of the picture space and atmosphere。