AI innovation in the supply chain of consumer packaged-goods for recognising objects in retail execution, supply chain management and smart factories: using novel diffusion-based optimisation algorithms and diffusion-based generative models

消费包装商品供应链中的人工智能创新，用于识别零售执行、供应链管理和智能工厂中的对象：使用新颖的基于扩散的优化算法和基于扩散的生成模型

• 批准号: 10081810
• 金额: $ 63.7万
• 依托单位: NEUROLABORATORIES LIMITED
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=10081810
• 关键词: AI; innovation; supply; chain; consumer; AI; innovation; supply; chain; consumer

Retail is a stepping stone on our path to becoming _a "plug-and-play" Computer Vision platform propelled by a synthetic data generation engine_ that can be adapted to multiple use cases & industries. In the future, we strive to become _a world-leading provider of computer vision as a service_ to the many use cases that exist across the supply chain of consumer product goods, i.e product sorting in waste management and recycling, product recognition in autonomous checkout stores, or inventory visibility in warehouses.Using the synergy between a team of experts at the University of Edinburgh and the South Korean team, we will be the pioneers in the creation of computer vision models that offer absolute guarantees for their performance. Our primary objective is to attain cross-domain generalisability for our platform, enabling us to streamline model development across various domains by harnessing our expertise in synthetic data generation. This approach will result in fewer models required to serve multiple clients efficiently.At Neurolabs, we are committed to democratising the field of computer vision, making sure that every small and medium-sized enterprise can harness the power of this transformative technology. To achieve this, we are open sourcing a selection of the groundbreaking algorithms developed throughout this project, incorporating them into optimisation packages. This strategic move aims to expand the applicability of these algorithms beyond Neurolabs' proprietary models, fostering their adoption across a wider range of applications.

零售是我们成为_a“插件”计算机视觉平台的道路上的垫脚石，该平台由合成数据生成引擎_推动，可以适应多种用例和行业。 In the future, we strive to become _a world-leading provider of computer vision as a service_ to the many use cases that exist across the supply chain of consumer product goods, i.e product sorting in waste management and recycling, product recognition in autonomous checkout stores, or inventory visibility in warehouses.Using the synergy between a team of experts at the University of Edinburgh and the South Korean team, we will be the pioneers in the creation为其性能提供绝对保证的计算机视觉模型。我们的主要目的是为我们的平台获得跨域的通用性，使我们能够通过利用合成数据生成方面的专业知识来简化各个领域的模型开发。这种方法将导致有效地为多个客户提供服务所需的模型。为了实现这一目标，我们正在开发整个项目中开发的开创性算法的选择，并将其纳入优化软件包。这一战略举动旨在将这些算法的适用性扩展到神经映射的专有模型之外，从而促进它们在更广泛的应用程序中的采用。