Computer Vision

A study using CT Foundation embeddings achieved a Receiver Operating Characteristic Area Under the Curve (AUC) of 0.791 for predicting distant metastasis risk in head and neck cancer, outperforming radiomics (AUC 0.772) …

The big-and-bulky last-mile delivery market is forecast to grow at roughly 5% annually through 2027, down from over 10% in the prior eight years, according to a joint report from Armstrong & Associates and the National H…

University of Michigan scientists used artificial intelligence to analyze over 170,000 skeletal measurements from more than 2,000 bird species, revealing that songbirds evolved through rare explosive bursts of change oft…

Roboflow has published a guide for building a machine-vision workflow that uses a custom-trained RF-DETR model to assess surface defect severity on leather, automatically classifying each detection as PASS, REVIEW, or FA…

Pico, an open-source AI color grading tool built on Claude, standardizes footage from S-Log3 to Rec.709 color space without automating the creative process. Developer Perbhat Bhatia positions Pico as a tool that removes …

Microsoft is rolling out a new OneDrive Photos app to some Windows 11 PCs that automatically indexes local images and adds AI-powered search and optional facial grouping, according to industry publication Windows Latest.…

Reflex launched Glass, an always-on screen recording app that uses secure, local inference on Apple's Neural Engine to provide visual context for AI agents. The app aims to enrich personal context by capturing screen act…

A developer built an autonomous, vision-capable crypto trading bot using Python 3.13 after studying 77 academic papers. The bot uses Google Gemini 3.6 Flash for visual chart analysis, a custom JIT-compiled indicator engi…

Procore Technologies will acquire DroneDeploy for approximately $845 million in cash, the companies announced July 29, 2026. The deal aims to integrate DroneDeploy's aerial and ground reality capture and robotics into Pr…

Elio, a sensor startup founded by former Meta AR/VR engineers Nadav Grossinger and Nitay Romano, raised $21 million in Series A funding co-led by Innovation Endeavors and Xora, with participation from Kevin Weil and Scri…

Onton co-founders Zach Hudson and Alex Gunnarson announced Ontology 1 on July 29th, claiming the AI model is at least 2.7 times more accurate than leading e-commerce search engines, though they did not identify competito…

PLOS Computational Biology published SynAPSeg on July 29, an open-source dataset and Python framework for detecting and measuring synaptic puncta in fluorescence microscopy images. The authors report custom StarDist mode…

Tether Data's QVAC research group released two 460M-parameter vision-language models, VisionPsy-Nano-460M and VisionPsy-Nano-460M-Flash, on July 29th, providing open weights and mobile inference code under the Apache 2.0…

Tavus launched PAL Maker on July 29th, a no-code builder for its video-based AI agents called PALs (Personal Affective Links), allowing nontechnical users to create and deploy agents that see, hear, and converse with use…

Reve shipped a Chrome extension on Wednesday that lets users save images from the web directly into its AI creation and editing platform, turning the browser into an input layer for visual references. The Palo Alto devel…

Pangram Labs Inc. raised $9 million in a funding round led by Menlo Ventures to improve its AI detection software for text and images, bringing its total funding to nearly $13 million. The company claims its Pangram 4 de…

Flock Safety CEO Garrett Langley acknowledged that human errors combined with the company's AI-powered license plate readers led to a wrongful police stop of a journalist and his wife, and that similar incidents are happ…

CopySight has raised a $3 million seed round led by Mucker Capital to build an intellectual property clearance layer for generative AI video, launching CopyScore V2 that scores AI-generated video for trademark, character…

AI has made previously 'virtually impossible' tasks like bird photo identification nearly trivial, but at the cost of reliability, argues Mike James in a July 2026 I Programmer article. James notes that neural networks a…

Researchers from the University of California, Riverside, YouTube, and Google DeepMind have developed SAGA (Source Attribution of Generative AI Videos), an AI system that can identify whether a video was AI-generated and…