Computer Vision news and analysis on Web Pulse: 3399 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Researchers introduced a hierarchical model for ball action anticipation in football broadcast video, achieving 17.91% mAP on the SoccerNet benchmark. The system uses a GRU and Transformer decoder with input-conditioned …
Researchers developed FUSE, a Bayesian framework that quantifies aleatoric and epistemic uncertainty in vision-language models, achieving state-of-the-art uncertainty calibration for downstream applications like robotics…
The Hisense 85-inch S7 CanvasTV has returned to its lowest price of $1,997.99 on Amazon, a 20% discount. The TV features a Hi-Matte display, QLED color, 4K resolution, and a 144Hz refresh rate, designed to blend into liv…
Kelowna entrepreneur Ryan Gallagher developed Back Road Intel, a sensor-and-AI system that surveys back roads to help emergency planners assess route condition ahead of wildfire season. The system collects data on road s…
Nvidia driver updates may help address cumulative color and exposure drift in AI-generated video clips, where small brightness increases accumulate over 4-5 chained clips, causing noticeable skin-tone and lighting change…
A developer built an offline skin lesion segmenter for mobile devices by optimizing the Med-SAM model with CoreML and ONNX Runtime. The approach enables real-time, privacy-preserving segmentation on iOS and cross-platfor…
NVIDIA researchers introduced VILA-M3, a vision-language model enhanced with medical expert knowledge, at CVPR 2025. The model aims to improve AI understanding of medical images by incorporating domain-specific expertise…
NVIDIA researchers published a study in IEEE Transactions of Medical Imaging investigating whether gradient inversion attacks compromise federated learning safety, finding that such attacks can reconstruct training data …
Niixo launched moneasy, a budgeting app that avoids bank login by using Apple Vision Framework and Gemini 2.5 Flash for receipt OCR and Siri for voice entry. Transactions are stored via CloudKit sync without external ser…
Researchers Kumar and Varshney proposed a six-pillar framework to mitigate homogenizing bias in AI models for facial aesthetic evaluation, as detailed in a JMIR preprint. The framework addresses underrepresentation of el…
Anthropic developed sophisticated execution harnesses for Claude that enable multi-step process management, error recovery, and tool integration, offering a blueprint for advanced LLM agent orchestration. A separate anal…
A developer built a real-time sign language reader that runs entirely in a web browser using client-side processing, demonstrating privacy-preserving local AI. The open-source project trycua/cua provides infrastructure f…
A developer explains that a neural network is not magic but a mechanical process of weighted detectors stacked in layers, turning raw input into higher-level features to make a decision. The network starts with random we…
Google's reCAPTCHA now offers hand gesture verification, collecting hand landmark data from videos that are deleted after verification and never associated with user identity. The feature requires camera permission and p…
Arduino has introduced support for running local Vision Language Models (VLMs) on the Arduino UNO Q board via the new App Lab software, enabling developers to deploy generative AI on low-power devices. A demonstration pr…
Rundiffusion Photo Flux defeated AuraFlow 20.8 to 15.5 in a head-to-head comparison, winning due to superior performance on the two hardest prompt-following tests: a ferry timetable poster and a macro scooter dashboard. …
Roboflow published a guide on integrating large language models into computer vision pipelines, detailing when to add an LLM for tasks like text extraction and contextual judgment versus when to rely on specialized detec…
MiniMax launched Hub, an all-in-one AI video generator platform, at the Shanghai International Film Festival. The tool consolidates image, video, voiceover, music, and editing into a single interface with a human-in-the-…
The 2026 FIFA World Cup features unprecedented technology integration, including motion-sensor balls, computer-vision player tracking, AI analysis for coaches and referees, cooling vests, upgraded telecom systems, and AI…
FastVideo released FastWan-QAD, a family of video generation models that can produce a 5-second 480P video in 1.78 seconds on a single NVIDIA RTX 5090 using quantization-aware distillation. The models achieve this speed …