My painting won an award despite being drawn via AI
This is a piece of work "Space Opera" that just won the first prize in the art competition held by the Colorado State Fair in the United States. The creator of the painting, Jason Allen, revealed that he used the AI painting app "Midjourney" to create the work.
The emergence of "Space Opera" has once again focused people's attention on the integration of AI and art. With the improvement of technology and software capabilities, the era of "everyone can draw" has arrived. The person in charge of Baidu Wenxin Yige said that the integration of computers and artistic creation has a lot of room for imagination, and AI will play a positive role in promoting artistic innovation.
Every ordinary person can describe his or her thoughts through words, and then create through machines. The learning and creation capabilities of artificial intelligence in the field of art are booming and constantly refreshing our cognition. New art categories are also emerging.
Is art dying because of AI?
"We are witnessing the death of art before our eyes." This is the most intuitive view of many netizens after learning about the creation process of "Space Opera". They believed that Allen's use of AI to generate images was completely cheating, but Allen himself disagreed.
He introduced that the creation process of this work took a month, constantly modifying and inputting keywords, and submitting as precise and specific words as possible to MidJourney. After creating more than 100 works, he selected three What Zhang himself found most satisfying was using GipaPixel AI to clarify the image, and then using Photoshop to make multiple adjustments before finally printing it on canvas.
This is not the first time that AI works have appeared on the official art stage. In October 2018, the art auction house Christie's sold an AI-generated oil painting " Portrait of Edmond Belamy. The creator asked the AI model to analyze 15,000 oil paintings from the 14th to 20th centuries, and then imitated a work with a similar style.
Christie’s describes the work as: an experiment in the interaction between art and artificial intelligence. This means that the core value of this auction work is not in the painting itself. People are paying more attention to the production method that seemed unprecedented at the time. This is enough to show that the pictures generated by AI technology are not a patchwork of existing pictures, but are similar to "re-creations" of humans after long-term painting training.
Although artificial intelligence works are full of impact and visual tension, like almost all other deep learning models, they are not good enough at understanding knowledge, reasoning, and logic. One problem that has been widely noticed is that artificial intelligence often generates weird-shaped hands.
This is where we first mention the principles of AI drawing. According to the relevant person in charge of Baidu Wenxin Yige, AI has learned the relationship between "linguistic description" and "artistic picture" from the massive data corresponding to pictures and texts, starting from a randomized starting point, constantly revising the paintings, and finally forming an aesthetic A finished product that is highly consistent with human experience and knowledge.
A human hand has more than 20 joints and is one of the richest structures in the human body. In most of the pictures used for training, the hand is often not the most central part. Different angles, different distances, different hand gestures, and being blocked by shadows and other objects will cause AI to have problems with the performance of the hand. question.
#From a technical point of view, there are two main aspects that need to be improved, including the controllability of generation and the ability to describe details. Currently, AI has no effective solutions to problems such as quantity, logic, relationship, and multi-graph association. For example, it is required to generate 2 apples, the left one is red and the right one is green. Although the relationship here is not responsible, under normal circumstances, the model cannot stably generate correct results.
On the other hand, AI’s ability to describe more complex and regular details needs to be improved. For example, the windows of a residential building should have multiple shapes, such as open, closed, someone watching the scenery by the window, clothes hanging on the balcony, etc... However, the current AI painting does not yet have such capabilities. Yes, hand painting also has the same problem.
The big model of "Strength can produce miracles"
The last time artificial intelligence surprised the Internet was in 2020.
This year, GPT-3 developed by OpenAI was born. A developer conducted a Turing test on GPT-3 and found that GPT-3 answered questions fluently, not like a machine. And ChatGPT, which has once again shocked the Internet recently, also comes from this company.
По сравнению с GPT-3, основным улучшением ChatGPT является его память. ChatGPT может хранить информацию о разговоре и продолжать контекст для обеспечения непрерывного диалога, что имеет решающее значение в сценариях разговора и значительно улучшает взаимодействие с пользователем в режиме взаимодействия с разговором.
ChatGPT неотделим от больших моделей, больших данных и больших вычислительных мощностей. Согласно данным исследовательского отчета Guosheng Securities, технологическая база ChatGPT использует точно настроенную модель серии GPT-3.5, которая имеет до 175 миллиардов параметров модели.Набор данных общедоступного сканера, в основном используемый OpenAI, содержит набор данных на человеческом языке, включающий более чем один триллион слов. . GPT-3.5 обучается на суперкомпьютерной инфраструктуре Microsoft Azure AI, а общее потребление вычислительной мощности составляет около 3640PF-дней (то есть один квадриллион вычислений в секунду в течение 3640 полных дней).
Ключевой технологией, обеспечивающей высококачественное применение искусственного интеллекта для рисования, являются большие модели. В марте 2021 года OpenAI выпустила продукт для рисования на основе искусственного интеллекта DALL·E, который был обновлен до DALL·E 2; интернет-гиганты, такие как Google, Meta, Baidu и Alibaba, также выпустили крупные модели искусственного интеллекта и соответствующие продукты, поддерживающие генерацию контента. .
Baidu запустила Wenxin Yige, платформу для создания искусственного интеллекта и вспомогательных инструментов, основанную на модели Wenxin. Она основана на постоянных технологических инновациях Baidu в области моделей Fei Paddle и Wenxin и выпущена в аспекте «ИИ-рисования». технологические инструменты. Предназначенный для людей с дизайнерскими потребностями и творческими способностями, он интеллектуально генерирует разнообразные творческие изображения искусственного интеллекта на основе большой модели Wenxin, чтобы помочь в творческом дизайне и устранить творческие узкие места.
Wenxin Yige — это инновационный продукт, реализованный с помощью системы диаграмм Wensheng на основе большой модели Wenxin. На официальном сайте Wenxin Yige пользователям нужно всего лишь ввести свой творческий текст и выбрать желаемый стиль рисования, чтобы быстро получить соответствующие картины, созданные Yige. Wenxin Yige может создавать не только великолепные и великолепные художественные картины, но также креативные и гиперреалистичные изображения с китайской спецификой. Он поддерживает более десяти стилей живописи и различных форматов, таких как китайский стиль, анимация, иллюстрация и живопись маслом.
Baidu Wenxin Large Model исследовала путь к индустриализации больших моделей. Технический директор Baidu Ван Хайфэн отметил, что компании, обладающие обширными преимуществами в алгоритмах, вычислительной мощности и данных, могут инкапсулировать сложный процесс производства моделей и предоставлять масштабные модельные услуги тысячам отраслей с помощью низкопороговых и высокоэффективных производственных платформ, тем самым формируя большую Масштабная модель пути индустриализации.
Прорыв границ живописи
В 1839 году появился фотоаппарат, оказавший огромное влияние на чисто ручную живопись того времени. Известный французский художник Деларош также сетовал: «Живопись мертва!» Однако с ростом популярности фотоаппаратов концепция фотографии как произведения искусства постепенно стала восприниматься публикой.
Во время знаменитого «Луддитского движения» в Британии в 19 веке большое количество британских рабочих ворвались на фабрики и уничтожили текстильные и другие машины, потому что считали, что машины отняли у них рабочие места.
Но, в конце концов, живопись не умерла, в искусстве появились новые формы выражения, а машины помогли обществу повысить производительность. Экономист Кейнс однажды проанализировал, что появление новых технологий приведет к технической безработице. Чтобы избежать исключения, людям всегда приходится совершенствовать свои способности в применении различных технологий.
Судя по нынешней практике, рисование с помощью искусственного интеллекта привнесло новые творческие методы и значительно повысило эффективность рисования. Будь то профессиональные создатели визуального контента, такие как художники, дизайнеры и художники, или создатели текстового контента, такие как средства массовой информации и авторы, живопись с использованием ИИ, несомненно, предлагает творческие идеи и помогает в художественном творчестве, повышая как эффективность, так и качество.
В будущем технологии и приложения будут быстро развиваться. С технической точки зрения, в будущем ИИ сможет не только рисовать, но и генерировать оригинальные короткие видеоролики ИИ на основе текстового контента одним щелчком мыши, а также автоматически добавлять субтитры, голос, музыку и т. д. Например, на конференции Baidu Mobile Ecology во Вьентьяне в 2022 году «Звезда», написанная, сочиненная и спетая цифровой личностью Ду Сяосяо, запущенная приложением Baidu, станет первым клипом, полностью созданным AIGC. Часть изображения написана Вэнь Синь Иге. на основе текста. Создается автоматически. С точки зрения применения, рисование с помощью ИИ будет по-прежнему интегрироваться в различные сценарии создания контента. Помимо общих областей иллюстраторов и художественного дизайна, оно будет все более широко использоваться в производстве мобильного контента, играх, промышленном дизайне, образовании в области искусственного интеллекта и других сферах. сценарии в будущем.
Выдающихся художников невозможно заменить ИИ. Но чтобы быть более конкурентоспособными, обычным художникам, возможно, придется научиться использовать ИИ для художественного творчества, точно так же, как сегодняшние художники должны освоить программное обеспечение для цифрового рисования.
Как заявил в ответ на полемику создатель «Космической оперы»: «Я не буду за это извиняться, я не нарушал никаких правил и победил».
The above is the detailed content of My painting won an award despite being drawn via AI. For more information, please follow other related articles on the PHP Chinese website!
Hot AI Tools
Undresser.AI Undress
AI-powered app for creating realistic nude photos
AI Clothes Remover
Online AI tool for removing clothes from photos.
Undress AI Tool
Undress images for free
Clothoff.io
AI clothes remover
AI Hentai Generator
Generate AI Hentai for free.
Hot Article
Hot Tools
Notepad++7.3.1
Easy-to-use and free code editor
SublimeText3 Chinese version
Chinese version, very easy to use
Zend Studio 13.0.1
Powerful PHP integrated development environment
Dreamweaver CS6
Visual web development tools
SublimeText3 Mac version
God-level code editing software (SublimeText3)
Hot Topics
1382
52
Bytedance Cutting launches SVIP super membership: 499 yuan for continuous annual subscription, providing a variety of AI functions
Jun 28, 2024 am 03:51 AM
This site reported on June 27 that Jianying is a video editing software developed by FaceMeng Technology, a subsidiary of ByteDance. It relies on the Douyin platform and basically produces short video content for users of the platform. It is compatible with iOS, Android, and Windows. , MacOS and other operating systems. Jianying officially announced the upgrade of its membership system and launched a new SVIP, which includes a variety of AI black technologies, such as intelligent translation, intelligent highlighting, intelligent packaging, digital human synthesis, etc. In terms of price, the monthly fee for clipping SVIP is 79 yuan, the annual fee is 599 yuan (note on this site: equivalent to 49.9 yuan per month), the continuous monthly subscription is 59 yuan per month, and the continuous annual subscription is 499 yuan per year (equivalent to 41.6 yuan per month) . In addition, the cut official also stated that in order to improve the user experience, those who have subscribed to the original VIP
Context-augmented AI coding assistant using Rag and Sem-Rag
Jun 10, 2024 am 11:08 AM
Improve developer productivity, efficiency, and accuracy by incorporating retrieval-enhanced generation and semantic memory into AI coding assistants. Translated from EnhancingAICodingAssistantswithContextUsingRAGandSEM-RAG, author JanakiramMSV. While basic AI programming assistants are naturally helpful, they often fail to provide the most relevant and correct code suggestions because they rely on a general understanding of the software language and the most common patterns of writing software. The code generated by these coding assistants is suitable for solving the problems they are responsible for solving, but often does not conform to the coding standards, conventions and styles of the individual teams. This often results in suggestions that need to be modified or refined in order for the code to be accepted into the application
Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations
Jun 11, 2024 pm 03:57 PM
Large Language Models (LLMs) are trained on huge text databases, where they acquire large amounts of real-world knowledge. This knowledge is embedded into their parameters and can then be used when needed. The knowledge of these models is "reified" at the end of training. At the end of pre-training, the model actually stops learning. Align or fine-tune the model to learn how to leverage this knowledge and respond more naturally to user questions. But sometimes model knowledge is not enough, and although the model can access external content through RAG, it is considered beneficial to adapt the model to new domains through fine-tuning. This fine-tuning is performed using input from human annotators or other LLM creations, where the model encounters additional real-world knowledge and integrates it
Seven Cool GenAI & LLM Technical Interview Questions
Jun 07, 2024 am 10:06 AM
To learn more about AIGC, please visit: 51CTOAI.x Community https://www.51cto.com/aigc/Translator|Jingyan Reviewer|Chonglou is different from the traditional question bank that can be seen everywhere on the Internet. These questions It requires thinking outside the box. Large Language Models (LLMs) are increasingly important in the fields of data science, generative artificial intelligence (GenAI), and artificial intelligence. These complex algorithms enhance human skills and drive efficiency and innovation in many industries, becoming the key for companies to remain competitive. LLM has a wide range of applications. It can be used in fields such as natural language processing, text generation, speech recognition and recommendation systems. By learning from large amounts of data, LLM is able to generate text
Five schools of machine learning you don't know about
Jun 05, 2024 pm 08:51 PM
Machine learning is an important branch of artificial intelligence that gives computers the ability to learn from data and improve their capabilities without being explicitly programmed. Machine learning has a wide range of applications in various fields, from image recognition and natural language processing to recommendation systems and fraud detection, and it is changing the way we live. There are many different methods and theories in the field of machine learning, among which the five most influential methods are called the "Five Schools of Machine Learning". The five major schools are the symbolic school, the connectionist school, the evolutionary school, the Bayesian school and the analogy school. 1. Symbolism, also known as symbolism, emphasizes the use of symbols for logical reasoning and expression of knowledge. This school of thought believes that learning is a process of reverse deduction, through existing
To provide a new scientific and complex question answering benchmark and evaluation system for large models, UNSW, Argonne, University of Chicago and other institutions jointly launched the SciQAG framework
Jul 25, 2024 am 06:42 AM
Editor |ScienceAI Question Answering (QA) data set plays a vital role in promoting natural language processing (NLP) research. High-quality QA data sets can not only be used to fine-tune models, but also effectively evaluate the capabilities of large language models (LLM), especially the ability to understand and reason about scientific knowledge. Although there are currently many scientific QA data sets covering medicine, chemistry, biology and other fields, these data sets still have some shortcomings. First, the data form is relatively simple, most of which are multiple-choice questions. They are easy to evaluate, but limit the model's answer selection range and cannot fully test the model's ability to answer scientific questions. In contrast, open-ended Q&A
SOTA performance, Xiamen multi-modal protein-ligand affinity prediction AI method, combines molecular surface information for the first time
Jul 17, 2024 pm 06:37 PM
Editor | KX In the field of drug research and development, accurately and effectively predicting the binding affinity of proteins and ligands is crucial for drug screening and optimization. However, current studies do not take into account the important role of molecular surface information in protein-ligand interactions. Based on this, researchers from Xiamen University proposed a novel multi-modal feature extraction (MFE) framework, which for the first time combines information on protein surface, 3D structure and sequence, and uses a cross-attention mechanism to compare different modalities. feature alignment. Experimental results demonstrate that this method achieves state-of-the-art performance in predicting protein-ligand binding affinities. Furthermore, ablation studies demonstrate the effectiveness and necessity of protein surface information and multimodal feature alignment within this framework. Related research begins with "S
SK Hynix will display new AI-related products on August 6: 12-layer HBM3E, 321-high NAND, etc.
Aug 01, 2024 pm 09:40 PM
According to news from this site on August 1, SK Hynix released a blog post today (August 1), announcing that it will attend the Global Semiconductor Memory Summit FMS2024 to be held in Santa Clara, California, USA from August 6 to 8, showcasing many new technologies. generation product. Introduction to the Future Memory and Storage Summit (FutureMemoryandStorage), formerly the Flash Memory Summit (FlashMemorySummit) mainly for NAND suppliers, in the context of increasing attention to artificial intelligence technology, this year was renamed the Future Memory and Storage Summit (FutureMemoryandStorage) to invite DRAM and storage vendors and many more players. New product SK hynix launched last year
