Openai官方给出的示例视频

Openai官方给出的示例视频 Prompt: “Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes.”

相关推荐

封面图片

“OpenAI发布文字生成视频模型“Sora”,文字prompt生成,可以创建长达 60 秒的,其中包含高度详细的场景、

“OpenAI发布文字生成视频模型“Sora”,文字prompt生成视频,可以创建长达 60 秒的视频,其中包含高度详细的场景、复杂的摄像机运动以及充满活力情感的多个角色。” “DEMO Prompt: “Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes.”” https://twitter.com/OpenAI/status/1758192957386342435 https://openai.com/sora by huburr from ibuki帰宅…

封面图片

AIGC变天?OpenAI推出文生视频模型Sora 官方合集来了

AIGC变天?OpenAI推出文生视频模型Sora 官方视频合集来了 Prompt:  “Beautiful, snowy Tokyo city is bustling. The camera moves through the  bustling city street, following several people enjoying the beautiful  snowy weather and shopping at nearby stalls. Gorgeous sakura petals are  flying through the wind along with snowflakes.”美丽雪景中的东京城繁忙喧嚣。摄像机穿过繁忙的街道,跟随几个人享受美丽的雪景和在附近摊位购物。美丽的樱花瓣随风飘落,与雪花一起飞舞。Prompt:  “Several giant wooly mammoths approach treading through a snowy meadow,  their long wooly fur lightly blows in the wind as they walk, snow  covered trees and dramatic snow capped mountains in the distance, mid  afternoon light with wispy clouds and a sun high in the distance creates  a warm glow, the low camera view is stunning capturing the large furry  mammal with beautiful photography, depth of field.”几头巨大的羊毛猛犸象踏过雪地草甸,它们长长的羊毛皮毛在风中轻轻摆动,远处覆盖着雪的树木和戏剧性的雪顶山脉,午后的光线透过稀疏的云层和远处高悬的太阳创造出温暖的光芒,低角度的摄像机视角令人惊叹,以美妙的摄影技术捕捉这大型毛茸茸的哺乳动物,景深效果。Prompt:  “A movie trailer featuring the adventures of the 30 year old space man  wearing a red wool knitted motorcycle helmet, blue sky, salt desert,  cinematic style, shot on 35mm film, vivid colors.”一部电影预告,展现一个穿着红色羊毛编织摩托车头盔的30岁太空人的冒险,蓝天、盐沼、电影风格,使用35mm胶片拍摄,色彩鲜艳。Prompt: “A gorgeously rendered papercraft world of a coral reef, rife with colorful fish and sea creatures.”一个华丽渲染的纸艺世界珊瑚礁,充满了五彩缤纷的鱼类和海洋生物。Prompt:  “Animated scene features a close-up of a short fluffy monster kneeling  beside a melting red candle. the art style is 3d and realistic, with a  focus on lighting and texture. the mood of the painting is one of wonder  and curiosity, as the monster gazes at the flame with wide eyes and  open mouth. its pose and expression convey a sense of innocence and  playfulness, as if it is exploring the world around it for the first  time. the use of warm colors and dramatic lighting further enhances the  cozy atmosphere of the image.”动画场景特写一个短小毛茸茸的怪物跪在一根融化的红蜡烛旁。艺术风格为3D且逼真,专注于光照和质感。画面氛围充满了惊奇和好奇,当怪物睁大眼睛,张开嘴巴凝视着火焰时。它的姿势和表情传达出一种纯真和顽皮的感觉,仿佛它是第一次探索周围的世界。温暖的色彩和戏剧性的光照进一步增强了画面的舒适氛围。Prompt:  “A stylish woman walks down a Tokyo street filled with warm glowing  neon and animated city signage. she wears a black leather jacket, a long  red dress, and black boots, and carries a black purse. she wears  sunglasses and red lipstick. she walks confidently and casually. the  street is damp and reflective, creating a mirror effect of the colorful  lights. many pedestrians walk about.”一位时尚女士走在东京街头,街道充满了温暖的霓虹灯和动态的城市标志。她穿着黑色皮夹克,一条长红裙和黑色靴子,并携带一个黑色手提包。她戴着太阳镜和红色口红,自信而随意地行走。街道潮湿且能反射,创造出色彩斑斓灯光的镜面效果。许多行人来往。Prompt: A bicycle race on ocean with different animals as athletes riding the bicycles with drone camera view一场在海面上的自行车比赛,不同的动物作为运动员骑着自行车,无人机摄像机视角。Prompt:  A instructional cooking session for homemade gnocchi hosted by a  grandmother social media influencer set in a rustic Tuscan country  kitchen with cinematic lighting由一位祖母社交媒体影响者主持的家庭自制意大利面食教学课程,设置在具有电影级照明的乡村托斯卡纳厨房中。Prompt: Two golden retrievers podcasting on top of a mountain两只金毛寻回犬在山顶上播客。Prompt: a futuristic drone race at sunset on the planet mars在火星上日落时分的未来派无人机比赛。纽约时报相关报道:OpenAI 公布新技术Sora,能即时生成令人瞩目的视频OpenAI,   一家位于旧金山的创业公司,最近推出了一项名为Sora的新技术。这项技术能够基于简短描述即时生成视频,展示了如同从好莱坞电影中提取出来的画面。Sora的演示视频包括毛茸茸的猛犸象在雪地中奔跑、怪物凝视融化的蜡烛,以及仿佛通过飞跃城市拍摄的东京街景等短片。作为ChatGPT聊天机器人和静态图像生成器DALL-E的背后公司,OpenAI正与包括Runway和大型科技巨头如Google及Meta(Facebook和Instagram的母公司)等在内的多家公司竞争,以改进这种即时视频生成技术。这项技术不仅可以加速经验丰富的电影制作人的工作,也可能完全取代经验较少的数字艺术家。然而,这项技术也可能成为在线虚假信息的快速且低成本的制作方式,使得网络上区分真实内容变得更加困难。华盛顿大学专门研究人工智能的教授Oren Etzioni表示:“我非常担心这类技术会影响一些竞争激烈的选举。”OpenAI将其新系统命名为Sora,取自日语中“天空”的意思,象征着无限的创造潜力。公司目前还未向公众发布Sora,因为它还在努力了解系统的潜在危险。相反,OpenAI正在与一小群学者和其他外部研究人员共享这项技术,这些人将对其进行“红队”测试,寻找可能的滥用方式。OpenAI正在对系统生成的视频加上水印,以标识这些视频是由AI生成的。但该公司承认这些水印可以被移除,且可能难以被察觉。Sora生成视频的方式是响应简短的描述,尽管视频可能令人印象深刻,但并不总是完美的,可能会包含奇怪和不合逻辑的图像。例如,该系统最近生成了一个人正在吃饼干的视频但饼干从未变小。随着DALL-E、Midjourney等静态图像生成器在过去几年的快速改进,它们现在生成的图像几乎与照片无法区分。这使得在线识别虚假信息变得更加困难,许多数字艺术家抱怨说,这让他们更难找到工作。密歇根州的电影概念艺术家Reid Southen表示:“当Midjourney在2022年首次推出时,我们都笑称‘哦,那很可爱’。现在,人们因为Midjourney而失去工作。” ... PC版: 手机版:

封面图片

OpenAI Sora问世 通往AGI的又一个ChatGPT时刻

OpenAI Sora问世 通往AGI的又一个ChatGPT时刻 因为看完Sora你可能会发现,OpenAI自己可能要用它先杀死GPT-4了。每个人都可以创造自己的世界了先来看看Sora。人们一直期待GPT-5,但Sora带来的轰动不亚于一次GPT-5的发布。作为OpenAI 首推的文本转视频模型,Sora能够根据文本指令或静态图像生成长达 1分钟的视频,其中包含精细复杂的场景、生动的角色表情以及复杂的镜头运动。同时也接受现有视频扩展或填补缺失的帧。每条提示60秒的视频长度与Pika Labs的3秒、Meta Emu Video的4秒、和Runway公司Gen-2的18秒相比,妥妥地铁赢了。并且从官方发布的演示来看,无论从视频流畅度还是细节表现能力上,Sora的效果都相当惊艳。比如官推里这条14秒的东京雪景视频。提示词:Beautiful, snowy Tokyo city is bustling. The camera moves through the bustling city street, following several people enjoying the beautiful snowy weather and shopping at nearby stalls. Gorgeous sakura petals are flying through the wind along with snowflakes.“美丽的,被雪覆盖的东京正繁忙着。镜头穿过繁忙的城市街道,跟随着几个享受雪景和在附近摊位购物的人。美丽的樱花瓣随风飘落,与雪花一同飞舞。”穿着时尚的女性漫步在霓虹背景的东京街头,地面有积水倒影。对于五官和皮肤的刻画十分真实,特别是痘印和法令纹,细节让人惊叹。猛犸象自冰川雪原中缓步从来,背后升腾起雪雾。烛火旁纯真顽皮的3D动画小怪物,光影、表情和毛茸茸的细节满分:一名24岁女性的眼部特写,足以以假乱真。无人机视角的海浪拍打着Big Sur加瑞角海岸崖壁,落日洒下金色光芒。窗台上的花朵开放延时影像:民众们上街舞龙,庆祝中国农历春节。可爱小猫咪在线安抚起床气。夜晚街道跑着的快乐小狗。两艘微型海盗船在一杯咖啡里对峙。加州淘金时代罕见“历史影像”流出 像吗?Sora做的。目前Sora还在测试阶段,仅对部分评估人员、视觉艺术家、设计师和电影制作人们开放访问权限,拿到试用资格的人们已经开始想象力横飞。Sam Altman就转发网友用Sora制作的“金光灿灿动物园”视频,玩起了自己的“What”梗:他还邀请大家踊跃提出想用Sora制作视频的提示词,团队马上为大家生成,瞬间8千多条回复。网友脑洞大开,要看海洋生物的自行车公开赛。两只金毛在山上戴着耳机开播客。当然也不忘cue一下去向成谜的Ilya,要求生成一个“Ilya眼中的真实世界”。不过OpenAI也表示,虽然Sora 对自然语言有着深入的理解,能够准确洞悉提示词,生成表达丰富的内容,并可以创建多个镜头、保持角色和视觉风格的一致性,但仍不可避免地存在一些弱点。例如,它在准确模拟复杂场景的物理现象方面存在困难,也可能不理解特定的因果关系。比方说“一个人咬一口饼干后,饼干上可能并没有咬痕。”模型也可能混淆提示的空间细节,像是弄错左右。或者“在准确体现随时间发生的事件方面遇到困难,比如遵循特定的摄像机轨迹”。Sora也使用了DALL·E 3的recaptioning技术,该技术涉及为视觉训练数据生成高度描述性的标题。因此模型能够更忠实地按照用户在生成视频中的文本指令进行操作。它能够一次性生成整个视频,或是扩展已生成的视频使其变长。通过让模型一次性预见多帧,解决了即使主体暂时离开视线也能保持不变的挑战性问题。关于安全性,OpenAI表示正与错误信息、仇恨内容和偏见等领域的专家合作,对模型进行对抗性测试。同时也在开发帮助检测误导性内容的工具,识别视频是否由Sora生成。对于违反使用政策的文本提示,例如暴力、仇恨和侵犯他人知识产权等内容,将拒绝显示给用户。除此以外,为DALL·E 3产品构建的现有安全方法也同样适用于Sora。“尽管进行了广泛的研究和测试,我们仍无法预测人们将如何利用我们的技术,也无法预见人们如何滥用它。这就是为什么我们相信,从真实世界的用例中学习,是随时间构建越来越安全的AI系统的关键组成部分。”OpenAI对 Sora信心满满,认为这为模型理解和模拟真实世界奠定了基础,是“实现AGI的重要里程碑”。网友们也第n+1次纷纷哀悼起相关赛道的公司们:“OpenAI就是不能停止杀死创业公司。”“天哪,现在起我们要弄清什么是真的,什么是假的。”“我的工作没了。”“整个影像素材行业被血洗,安息吧。”能杀死GPT-4的世界模型?这不就是吗OpenAI一如既往没有给出很详细的技术说明,但一些只言片语已经足够让你浮想联翩。其中最吸引我们注意的第一个点,是对数据的处理。Sora是一个扩散模型(diffusion model),采用类似GPT的Transformer架构。而在解决训练中文本资料与视频数据之间的统一方面,OpenAI表示,他们在处理图像和视频数据时,把对它们进行分割后得到的最小单元,称为小块(patches),也就是对应LLM里的基本单元tokens。这是一个很重要的技术细节。把它作为模型处理的基本单元,使得深度学习算法能够更有效地处理各种视觉数据,涵盖不同的持续时间、分辨率和宽高比。从最终的震撼效果看,你很难不得出这样一个结论:对语言的理解能力,是可以迁移到对更多形态的数据的理解方法上去的。此前的Dalle-3的效果就被公认很大程度来自OpenAI在GPT上积累的领先N代的语言能力,哪怕是个图像为输出的模型,语言能力提升也是至关重要的。而今天的视频模型,同样如此。至于它是如何做到的,有不少行业内的专家给出了相同的猜测:它的训练数据里使用了游戏领域最前端的物理引擎Unreal Engine5,简单粗暴的理解,就是语言能力足够强大之后,它带来的泛化能力直接可以学习引擎生成的图像视频数据和它体现出的模式,然后还可以直接用学习来的,引擎最能理解的方式给这些利用了引擎的强大技术的视觉模型模块下指令,生成我们看到的逼真强大的对物理世界体现出“理解”的视频。基于这个猜测,OpenAI简短的介绍中的这句话似乎就更加重要了:“Sora 是能够理解和模拟现实世界的模型的基础,OpenAI相信这一功能将成为实现AGI的重要里程碑。”理解,现实,世界。这不就是人们总在争论的那个唯一有可能“干掉”GPT-4的世界模型。现在,OpenAI搞出来了它的雏形,摆在了你的面前。看起来,这个模型学会了关于 3D 几何形状和一致性的知识,而且并非OpenAI训练团队预先设定的,而是完全是通过观察大量数据自然而然地学会的。负责Sora训练的OpenAI科学家Tim Brooks表示,AGI将能够模拟物理世界,而Sora就是迈向这个方向的关键一步。显然,在OpenAI眼里,它不只是一个“文生视频模型”,而是更大的东西。我们如果尝试进一步给一个暴论,就是:语言是理解一切的基础,理解了视频后,世界模型才会到来。可能这才是比今天刷屏感慨“现实不存在了”之外,更恐怖的地方。这可能是人类通往AGI的又一个ChatGPT时刻。 ... PC版: 手机版:

封面图片

Perplexity 的官方 Prompt

Perplexity 的官方 Prompt # Assistant Background You are Perplexity, a helpful search assistant trained by Perplexity AI. # General Instructions Write an accurate, detailed, and comprehensive response to the user''s INITIAL_QUERY. Additional context is provided as "USER_INPUT" after specific questions. Your answer should be informed by the provided "Search results". Your answer must be precise, of high-quality, and written by an expert using an unbiased and journalistic tone. Your answer must be written in the same language as the question, even if language preference is different. You MUST cite the most relevant search results that answer the question. Do not mention any irrelevant results. You MUST ADHERE to the following instructions for citing search results: - to cite a search result, enclose its index located above the summary with brackets at the end of the corresponding sentence, for example "Ice is less dense than water." or "Paris is the capital of France." - NO SPACE between the last word and the citation, and ALWAYS use brackets. Only use this format to cite search results. NEVER include a References section at the end of your answer. - If you don't know the answer or the premise is incorrect, explain why. If the search results are empty or unhelpful, answer the question as well as you can with existing knowledge. You MUST NEVER use moralization or hedging language. AVOID using the following phrases: - "It is important to ..." - "It is inappropriate ..." - "It is subjective ..." You MUST ADHERE to the following formatting instructions: - Use markdown to format paragraphs, lists, tables, and quotes whenever possible. - Use headings level 2 and 3 to separate sections of your response, like "## Header", but NEVER start an answer with a heading or title of any kind. - Use single new lines for lists and double new lines for paragraphs. - Use markdown to render images given in the search results. - NEVER write URLs or links. # Query type specifications You must use different instructions to write your answer based on the type of the user's query. However, be sure to also follow the General Instructions, especially if the query doesn't match any of the defined types below. Here are the supported types. ## Academic Research You must provide long and detailed answers for academic research queries. Your answer should be formatted as a scientific write-up, with paragraphs and sections, using markdown and headings. ## Recent News You need to concisely summarize recent news events based on the provided search results, grouping them by topics. You MUST ALWAYS use lists and highlight the news title at the beginning of each list item. You MUST select news from diverse perspectives while also prioritizing trustworthy sources. If several search results mention the same news event, you must combine them and cite all of the search results. Prioritize more recent events, ensuring to compare timestamps. You MUST NEVER start your answer with a heading of any kind. ## Weather Your answer should be very short and only provide the weather forecast. If the search results do not contain relevant weather information, you must state that you don't have the answer. ## People You need to write a short biography for the person mentioned in the query. If search results refer to different people, you MUST describe each person individually and AVOID mixing their information together. NEVER start your answer with the person's name as a header. ## Coding You MUST use markdown code blocks to write code, specifying the language for syntax highlighting, for example ```bash or ```python If the user's query asks for code, you should write the code first and then explain it. ## Cooking Recipes You need to provide step-by-step cooking recipes, clearly specifying the ingredient, the amount, and precise instructions during each step. ## Translation If a user asks you to translate something, you must not cite any search results and should just provide the translation. ## Creative Writing

🔍 发送关键词来寻找群组、频道或视频。

启动SOSO机器人