search
HomeTechnology peripheralsAIMicrosoft AR/VR patent sharing provides users with personalized experiences and digital environment adaptation methods based on saliency

Saliency-based digital environment adaptation method

英伟网Nweon October 27, 2023) It is very difficult to create personalized experiences for users in the XR digital environment. Additionally, it is equally difficult to adapt a digital environment based on content related to another digital environment,

So in the patent application titled "Saliency-based digital environment adaptation", Microsoft introduced a saliency-based digital environment adaptation method. In examples, the digital environment may be adjusted based on a variety of factors, including content attributes, environment attributes, user profile attributes, and/or group attributes, among others. Accordingly, a salience measure may be determined for the content and/or location of the digital environment based on the factors described.

Next, content from a set of content can be determined based on an associated saliency measure, where the content set is ordered according to the saliency measure of each content instance.

Thus, a saliency measure for the highest ranking displayed to the user can be determined. For example, 2D or 3D assets can be presented to the user, and/or environment mechanics can be incorporated or modified

As another example, similar techniques can be leveraged to determine spatial locations for displaying content to users and to rank a set of spatial locations based on relevant saliency measures. Therefore, the experience provided by the digital environment may vary from user to user, providing a personalized experience for each user

Microsoft AR/VR patent sharing provides users with personalized experiences and digital environment adaptation methods based on saliency

To realize the function of adapting to the digital environment. Digital environment platform 102 may be a software application or a hardware device used to manage and control digital environment services 104 and computing devices 106 . The digital environment service 104 can provide various functions, including data storage, data processing, data analysis, user interface, etc. Computing device 106 may be a personal computer, smartphone, tablet, or other device connectable to a network. Through network 108, digital environment platform 102 may communicate with digital environment services 104 and computing devices 106 in order to achieve digital environment adaptation goals

Digital environment platform 102 can aggregate telemetry data related to one or more digital environments. The digital environment platform 102 includes a request processor 110, a saliency measurement engine 112, an interaction data store 114, and a content data store 116

In an example, request handler 110 handles various requests that may be received from digital environment services 104 and computing devices 106 . For example, request processor 110 may handle requests for saliency metrics associated with content. The request may contain an indication of the content whose significance is measured, an indication of the relevant digital environment, and/or one or more statistical data.

The saliency metric engine 112 may generate a saliency metric and/or an indication of location and/or content. For example, the saliency metric engine 112 may process telemetry data stored in the interaction data store 114 and/or content attributes related to the candidate content.

The saliency metric engine 112 may use any of a variety of techniques to generate saliency metrics and/or determine content and/or location from a set of candidates based on the correlation factors described above.

System 100 further includes digital environment services 104 that may be used to provide a digital environment. For example, when presenting a digital environment for display to a user of computing device 106, environment application 122 and digital environment service 104 may run as a client and server, respectively.

In other examples, environment application 122 may run locally such that digital environment service 104 may distribute environment application 122 to any of a variety of computing devices.

Digital environment service 104 consists of saliency processor 118 and content data storage 120. In an example, saliency processor 118 is used to generate and/or obtain telemetry data

Additionally, the saliency processor 118 may request saliency metrics from the digital environment platform. For example, saliency processor 118 may request a saliency measure for content from content data store 120 and/or may request external content from digital environment platform 102 .

As described above, environment application 122 may generate a digital environment for presentation to a user of computing device 106 . As another example, at least a portion of the digital environment may be presented through digital environment service 104.

Accordingly, saliency processor 124 and/or saliency processor 118 may determine content for adaptation to the digital environment. For example, saliency processor 124 may request saliency metrics and/or content from digital environment platform 102 .

In one case, the request includes at least a portion of a user profile, which may be stored by computing device 106. In other cases, information may be stored via digital environment services 104 and/or digital environment platform 102

The role of the environment application 122 is to determine the spatial location of the content, select the content to be displayed to the user, and adjust the environment mechanism based on the determined content

In other examples, any number of computing devices can be used. In these examples, the digital environment can be adapted to the user of each computing device to present different relevant representations to each user

For example, a first user may view the digital environment as containing a first content item, while a second user may view the digital environment as containing a second content item. Similarly, different environment mechanisms can be suitable for different users, just like the first user likes a certain environment mechanism, and the second user does not like this environment mechanism

As a further example, external content may be presented to a first user based on a first interest set associated with the first user, while external content may be presented to a second user based on a second interest set associated with the second user. Two users rendering external content

Microsoft AR/VR patent sharing provides users with personalized experiences and digital environment adaptation methods based on saliency

Figure 2 illustrates an example method 200 for generating a content significance measure.

Starting from operation 202, we can obtain a set of content attributes. For example, these attribute sets could include the relative accessibility and/or scarcity of the content

Operation 204, obtain a set of environment attributes. In an example, the set of environment attributes relates to a spatial location in the digital environment where the content may be presented, such as a spatial location close to the user. As another example, a set of environment attributes may include indications regarding the user's progress in the digital environment's storyline.

Operation 206, obtain a set of user profile attributes. For example, a user profile attribute set may be related to the user's gaming or interaction style, the user's attention habits (e.g., based on the user's perspective determined from an AR/VR headset), etc.

In operation 208, a set of attributes is obtained. In an example, the attribute set includes attributes similar to those obtained in operation 206, but aggregated based on one or more statistics. For example, a set of demographic attributes may be determined based on telemetry data associated with one or more digital environments (eg, may be provided by an interaction data store such as interaction data store 114). For example, demographic attributes may indicate difficulty levels and/or popularity related to game mechanics and/or the spatial location of the user and/or content that may be determined.

The content rewritten in Chinese is as follows: Operation 210 is to generate a saliency measure of the content based on the attributes obtained from operations 202-208. Different aspects of operation 210 may include generating a saliency measure using a machine learning model. As an example, operation 210 might generate a saliency measure based on a set of weights associated with each attribute

Operation 212 provides an indication of the generated saliency measure. For example, a significance metric may be provided in response to a significance metric request. As another example, a saliency measure can be provided and used to rank a set of content.

Microsoft AR/VR patent sharing provides users with personalized experiences and digital environment adaptation methods based on saliency

Figure 3 illustrates an example method 300 for adjusting a digital environment.

After starting operation 302, a set of candidate content can be obtained. For example, some of this content may be related to the digital environment. Additionally, some of this content may include content from external sources. These content collections are available from a variety of sources

Operation 304, determine a saliency measure for the content set. For example, operation 304 may include generating a saliency measure for each content instance in the content set obtained at operation 302.

Based on the significance measure generated by operation 304, the content set is sorted and proceeds to operation 308. Content is determined from the sorted content set. For example, one or more top-ranked content instances can be selected, or content can be randomly selected from the content instances based on a significance measure that is above a preset threshold.

Operation 310, adjust the digital environment according to the content determined in operation 308. For example, operation 310 may include adapting the digital environment to include 2D or 3D assets or NPCs for presentation to the user. Another example would be to present users with different RPG story options, either with varying difficulty, prominence, and/or length.

Microsoft AR/VR patent sharing provides users with personalized experiences and digital environment adaptation methods based on saliency

Another example method of adjusting the digital environment is shown in Figure 4 400

Starting from operation 402, a set of locations is determined. In an example, the location set may be determined based on the user's location in the digital environment. For example, a location set may include surfaces that are close to the user.

For the rewritten content, we need to convert it to Chinese and rewrite it without changing the original meaning

What needs to be rewritten is: operation 406, sorting the location set according to the significance measure generated in operation 404. Moving to operation 408, the location is determined based on the ordered set of locations. For example, one or more top-ranked positions may be selected, or as another example, positions may be randomly selected from among positions that have a significance measure above a preset threshold

Operation 410, determine content to adapt to the digital environment. Operation 412 uses the content determined at operation 410 to adjust the digital environment based on the location determined at operation 408. For example, operation 412 may include adjusting the digital environment to include a 2D or 3D asset or NPC for presentation to the user at a determined location.

Related Patents: Microsoft Patent | Saliency-based digital environment adaptation
https://patent.nweon.com/30770

The Microsoft patent application titled "Saliency-based digital environment adaptation" was originally submitted in March 2022 and was recently published by the US Patent and Trademark Office.

---

Original link: https://news.nweon.com/114274 The content that needs to be rewritten is: Original link: https://news.nweon.com/114274

The above is the detailed content of Microsoft AR/VR patent sharing provides users with personalized experiences and digital environment adaptation methods based on saliency. For more information, please follow other related articles on the PHP Chinese website!

Statement
This article is reproduced at:搜狐. If there is any infringement, please contact admin@php.cn delete
微软深化与 Meta 的 AI 及 PyTorch 合作微软深化与 Meta 的 AI 及 PyTorch 合作Apr 09, 2023 pm 05:21 PM

微软宣布进一步扩展和 Meta 的 AI 合作伙伴关系,Meta 已选择 Azure 作为战略性云供应商,以帮助加速 AI 研发。在 2017 年,微软和 Meta(彼时还被称为 Facebook)共同发起了 ONNX(即 Open Neural Network Exchange),一个开放的深度学习开发工具生态系统,旨在让开发者能够在不同的 AI 框架之间移动深度学习模型。2018 年,微软宣布开源了 ONNX Runtime —— ONNX 格式模型的推理引擎。作为此次深化合作的一部分,Me

微软提出自动化神经网络训练剪枝框架OTO,一站式获得高性能轻量化模型微软提出自动化神经网络训练剪枝框架OTO,一站式获得高性能轻量化模型Apr 04, 2023 pm 12:50 PM

OTO 是业内首个自动化、一站式、用户友好且通用的神经网络训练与结构压缩框架。 在人工智能时代,如何部署和维护神经网络是产品化的关键问题考虑到节省运算成本,同时尽可能小地损失模型性能,压缩神经网络成为了 DNN 产品化的关键之一。DNN 压缩通常来说有三种方式,剪枝,知识蒸馏和量化。剪枝旨在识别并去除冗余结构,给 DNN 瘦身的同时尽可能地保持模型性能,是最为通用且有效的压缩方法。三种方法通常来讲可以相辅相成,共同作用来达到最佳的压缩效果。然而现存的剪枝方法大都只针对特定模型,特定任务,且需要很

超5800亿美元!微软谷歌神仙打架,让英伟达市值飙升,约为5个英特尔超5800亿美元!微软谷歌神仙打架,让英伟达市值飙升,约为5个英特尔Apr 11, 2023 pm 04:31 PM

ChatGPT在手,有问必答。你可知,与它每次对话的计算成本简直让人泪目。此前,分析师称ChatGPT回复一次,需要2美分。要知道,人工智能聊天机器人所需的算力背后烧的可是GPU。这恰恰让像英伟达这样的芯片公司豪赚了一把。2月23日,英伟达股价飙升,使其市值增加了700多亿美元,总市值超5800亿美元,大约是英特尔的5倍。在英伟达之外,AMD可以称得上是图形处理器行业的第二大厂商,市场份额约为20%。而英特尔持有不到1%的市场份额。ChatGPT在跑,英伟达在赚随着ChatGPT解锁潜在的应用案

竞争加剧,微软和Adobe发布AI图像生成工具竞争加剧,微软和Adobe发布AI图像生成工具Apr 11, 2023 pm 10:55 PM

随着OpenAI DALL-E和Midjourney的推出,AI艺术生成器开始变得越来越流行,它们接受文本提示并将其变成美丽的、通常是超现实的艺术品——如今,有两家大企业加入了这一行列。微软宣布,将通过Bing Image Creator把由DALL-E模型提供支持的AI图像生成功能引入Bing搜索引擎和Edge浏览器。创意软件开发商Adobe也透露,将通过名为Firefly的AI艺术生成产品来增强自己的工具。对于有权访问Bing聊天预览的用户来说,这一新的AI图像生成器已经可以在“创意”模式下

微软推出AI工具Security Copilot,帮助网络安全人员应对威胁微软推出AI工具Security Copilot,帮助网络安全人员应对威胁Apr 04, 2023 pm 02:50 PM

近日微软推出了Security Copilot,这款新工具旨在通过AI助手简化网络安全人员的工作,帮助他们应对安全威胁。 网络安全人员往往要管理很多工具,和来自多个来源的海量数据。近日微软宣布推出了Security Copilot,这款新工具旨在通过AI助手简化网络安全人员的工作,帮助他们应对安全威胁。Copilot利用基于OpenAI的GPT-4最新技术,让网络安全人员能够就当前影响环境的安全问题提问并获得答案,甚至可以直接整合公司内部的知识,为团队提供有用的信息,从现有信息中进行学习,将当前

NLP模型读不懂人话?微软AdaTest挑错效率高五倍NLP模型读不懂人话?微软AdaTest挑错效率高五倍Apr 09, 2023 pm 04:11 PM

​自然语言处理(NLP)模型读不懂人话、将文本理解为相反的意思,是业界顽疾了。 现在微软表示,开发出解决此弊的方法。微软开发AdaTest方法来测试NLP模型 可作为跨越各种应用基础的大型模型,或称平台模型的进展已经大大改善了AI处理自然语言的能力。但自然语言处理(NLP)模型仍然远不完美,有时会以令人尴尬的方式暴露缺陷。 例如有个顶级的商用模型,将葡萄牙语中的「我不推荐这道菜」翻译成英语中的「我非常推荐这道菜」。 这些失败之所以继续存在,部分原因是寻找和修复NLP模型中的错误很难,以至于严重的

微软出品的Python小白神器,真香!微软出品的Python小白神器,真香!Apr 12, 2023 am 10:55 AM

大家好,我是菜鸟哥!最近逛G网,发现微软开源了一个项目叫「playwright-python」,作为一个兴起项目。Playwright 是针对 Python 语言的纯自动化工具,它可以通过单个API自动执行 Chromium,Firefox 和 WebKit 浏览器,连代码都不用写,就能实现自动化功能。虽然测试工具 selenium 具有完备的文档,但是其学习成本让一众小白们望而却步,对比之下 playwright-python 简直是小白们的神器。Playwright真的适用于Python吗?

微软必应再强化!接入OpenAI DALL·E模型,文字生成图像微软必应再强化!接入OpenAI DALL·E模型,文字生成图像Mar 31, 2023 pm 10:39 PM

微软必应完善文字生成图像能力,Adobe 今日也发布 Firefly,杀入生成式 AI 这场游戏。 昨晚实在是有些热闹。一边英伟达 GTC 正在进行中,一边谷歌正式开放了 Bard 的测试,这里微软必应也不甘寂寞。今日,微软正式宣布,必应搜索引擎接入了 OpenAI 的 DALL·E 模型,增加了 AI 生成图像的功能。也就是说,在接入 ChatGPT 之后,必应再次强化,Bing Image Creator 能够让用户用 DALL·E 模型生成图像。「对于拥有必应预览版权限的用户,Bing I

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

AI Hentai Generator

AI Hentai Generator

Generate AI Hentai for free.

Hot Tools

SublimeText3 English version

SublimeText3 English version

Recommended: Win version, supports code prompts!

SAP NetWeaver Server Adapter for Eclipse

SAP NetWeaver Server Adapter for Eclipse

Integrate Eclipse with SAP NetWeaver application server.

WebStorm Mac version

WebStorm Mac version

Useful JavaScript development tools

SublimeText3 Linux new version

SublimeText3 Linux new version

SublimeText3 Linux latest version

MinGW - Minimalist GNU for Windows

MinGW - Minimalist GNU for Windows

This project is in the process of being migrated to osdn.net/projects/mingw, you can continue to follow us there. MinGW: A native Windows port of the GNU Compiler Collection (GCC), freely distributable import libraries and header files for building native Windows applications; includes extensions to the MSVC runtime to support C99 functionality. All MinGW software can run on 64-bit Windows platforms.