Polycam

Docsumo is a document processing tool driven by AI that automates the extraction of data from diverse unstructured documents like invoices, financial statements, and contracts. It leverages Intelligent OCR and machine learning to precisely capture text and data from scanned documents, PDFs, and images. Various industries might adopt Docsumo to enhance efficiency, minimize errors, save time and costs tied to manual data entry, and simplify their document processing workflows. The software's adaptability, scalability, and integration capabilities make it appealing for organizations seeking to automate and enhance their document management processes.

Segment Anything AI (Meta) provides the Segment Anything Model (SAM), an AI tool capable of isolating any object within any image. SAM is promptable and exhibits zero-shot generalization to novel images and objects, utilizing a range of input prompts that allow seamless integration with other AI systems. It can also be trained to label images and enhance its dataset. The SAM model is crafted to be efficient and adaptable, optimizing its data engine's performance. Contributors to the project include Alexander Kirillov, Eric Mintun, Nikhila Ravi, among others. The code is accessible on GitHub, and users can subscribe to their newsletter for updates on their latest research advancements.

Polycam is a smartphone app for 3D capture that allows users to produce high-quality 3D models from photos using any device and quickly scan spaces with a LiDAR sensor. It also provides the ability to edit 3D captures directly on the device and export them in multiple file formats, share them with friends and the Polycam community, and take measurements and create blueprints.