Original title: Everyone makes a film and television drama that they are satisfied with, but Sora can't do it?
Read it "Romance of the Three Kingdoms" and try it", ""The Three-Body Problem" can be directly transferred to a movie", ......Recently, OpenAI released the first Wensheng ** model Sora, which quickly swiped the screen on the Internet, and many netizens were eager to try it, looking forward to using AI to restore the ** scene in the future.
There are also people**, everyone will be able to make film and television dramas that they are satisfied with in the future. Will this be possible in the future? Or how far are we from achieving this?
Feed ** into AI, and then what?
Before talking about SORA, I have to mention that many people in the film and television game industry have already applied AIGC (artificial intelligence generated content) to content production.
Mr. Feng ("AI Maniahlum"), a blogger from Sichuan, has worked in the CG field for 15 years, engaged in film and television special effects post-production, game development, etc. He said that the film and television industry is now widely accepting the entry of AI, which is currently mainly used in the concept design and stage design of film and television in the early stage, and some animation teams and game development companies are also laying out or introducing AI production lines.
Mr. Feng recently released a number of AI concept animations of the "Journey to the West" series on the short ** platform, which has received nearly one million ** volumes. He told reporters that the tools he uses are the AI painting tool Midjourney and the AI generation tool Runway.
Each shot has to be conceived first, and then drawn through Midjourney, a shot may have to draw thousands, and finally select one, in the episode "The Monkey King" I drew three or four thousand, and finally selected a hundred or so shots, and then took these shots to the AI tool to generate animation, and then edited. The lines are designed first, and they are dubbed by AI, and if there are some special ones, I will dub them myself, and then use a voice changer to adjust the effect. ”
Mr. Feng said that it takes about a week to make such a section with AI, but it can take several months to make it manually. He said that in general, animation requires conceptual design, original painting, 2D frame-by-frame storyboarding, 3D scene construction and other processes, but if there is AI assistance, as long as there is the original painting, AI will help you produce the picture and generate animation effects, and the intermediate process will save a lot.
And the advent of SORA may save this process even more. In Mr. Feng's opinion, Sora is able to generate multiple shots in a single frame, which he can't do with the tools he used in the past, which means that he can create smoother and more complete works in the future.
So, can you get a satisfactory ** by inputting ** in?Mr. Feng believes that there is still a process to be done, and if it is fast, it may take three or four years. "Actually, it's not a technical problem, it's hard for humans to be able to understand the literature of different countries, understand its background and different cultural elements, but AI doesn't understand these things very well. ”
Mr. Feng mentioned that in the process of creating with AI tools, it can be clearly felt that it does not have a thorough understanding of different cultures, such as Chinese culture, which looks similar in style, but if you look closely at the patterns of clothes and armor, they are not real.
However, he also said that AI learning has grown exponentially, and it has only been iterated for more than a year from picture abstraction to being able to understand the richness and style of the picture. SORA does have flaws, but it's just an iterative process of AI, and it's not a big deal in the future.
Sora came out, and everyone became a director
Although SORA is not yet open to the public, many people believe that it is no less significant than the release of ChatGPT 1 year ago. From a technical point of view, SORA is strong in the world
Zhang Jinbao, an associate professor at Beijing Normal University's Faculty of Education, said that SORA uses a combination of diffusion models and DIFFUSION transformers (DITS). The diffusion model destroys the image by gradually adding noise, and then learns the reverse process to recover the image. The Transformer architecture captures global dependencies in the image.
The SORA generation process can be divided into the following steps: converting a text description into a series of semantic vectorsInput semantic vectors into the model to generate a latent representation of **Decode the latent representation into pixels, generating the final **.
Zhang Jinbao said that what we have seen so far is that SORA uses a large number of ** and image datasets for training, including movies, TV series, documentaries, game screens, etc. These datasets cover a variety of scenes, characters, and actions, providing rich learning material for the model.
In his view, the birth of SORA not only marks a major progress in generation technology, but also brings unprecedented opportunities and challenges to content creation, entertainment and other industries. For example, the emergence of SORA indicates that the competition in the field of generation will become more intense, driving the rapid development of multimodal AI and various AI application scenarios, especially in industries that require real-world modeling. Compared to traditional film and television production, the production process with SORA is simpler, the investment cost is lower, and the creative results are faster.
Zhang Jinbao believes that SORA allows creators to use AI tools more freely to express their ideas, reduces the constraints of industrial processes, and provides new perspectives and tools for content creation. Although SORA cannot completely replace traditional film and television production at present, its powerful ability shows that the direction of relying on AI to assist human creativity is becoming more and more clear, which may redefine the way film and television content is created and consumed.
This is also one of the reasons why AI has caused so much discussion, it will bring pressure to everyone, in fact, the practitioners around me are basically unaffected, if you are a mature and experienced artist or special effects, it will not be affected too much, it will become an auxiliary tool for you. Mr. Feng believes that even if there is AI, it is unrealistic for everyone to become a director, and in the end, it is still in the hands of a few professionals.
Ma Heliang, executive secretary general of the Science Fiction Film Working Committee of the China Film Association, also said in an interview that in the short term, positions related to concept design and preview will be affected by SORA to a certain extent, but the effect it can present is actually not up to the level of cinema films, and the film has a subjective position, perspective and expression in creation, and there are emotional exchanges and emotional projections from time to time, which is far from being replaced by simply generating a paragraph. Therefore, AI as a technical aid is indeed already changing the way and process of film production, and can optimize and improve film creation, but it is too early to say that it can "disrupt the entire film industry".
What are the other possibilities for AI?
Now that Pandora's box has been opened, it is unrealistic to expect it to close. In Mr. Feng's view, in addition to painting, doing **, copywriting, etc., AI has more room for development.
The SORA model demonstrates the ability to further imagine more application possibilities and bring new changes and innovations to various fields. For example, in the field of education, it can be used to create personalized teaching**, simulation experiments and scenarios, and be used to create virtual classrooms to provide more convenient educational resources.
In the medical field, it can be used to make medical animations to help doctors and patients better understand the condition; Simulate the surgical process to help doctors plan and train before surgery; Telemedicine is carried out to provide convenient medical services to patients in remote areas. In the field of scientific research, it can be used to simulate scientific experiments, generate scientific data visualizations, and build virtual worlds for scientific research and exploration. In the business sector, it can be used to create product promotions**, create virtual showrooms, use them for market research and analysis, help businesses better understand customer needs, and more.
At the same time, there is also a lot of concern about the risks that SORA may pose, such as the possibility of being used for counterfeiting or deception, and the content generated by SORA may be biased or incorrect. Legal professionals believe that with the development of AI technology, legal regulation also needs to keep up. Service providers or providers also need to take effective measures to prevent all kinds of discrimination in the process of algorithm design, training data selection, model generation and optimization, and provision of services, and at the same time avoid producing, copying, publishing, and disseminating false information.
In Zhang Jinbao's view, intelligence will fully penetrate human society, and people will learn to understand technology, capture the potential of emerging technologies, and promote the orderly and dynamic development of society in the process of widely using various intelligent technologies to enhance competitiveness and complete tasks. "Capturing new trends in technological development and making the right decisions within a controllable range is not just a test of one person, but the resilience of the whole society. (End) (Reporter Yuan Xiuyue).
*: China News Network].