Computer Vision news and analysis on Web Pulse: 1790 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Nader Al Shawki, a final-year AI Engineering student at Al-Razi University in Yemen, has built several production-grade AI systems including a tomato leaf disease detection platform using YOLOv8 and FastAPI, a pneumonia …
A developer tested four AI tools—ChatGPT, Claude, Grok, and Gemini—on creating a banner for a cricket academy. ChatGPT produced the best image, Grok made the best video, Claude defaulted to building a webpage instead of …
Apple's HomeKit Secure Video service is gaining AI-powered features in iOS 27, including descriptive alerts and natural-language search, making it more competitive with rivals like Ring and Google Nest. The improvements,…
Swsh, a college photo-sharing app turned AI-powered fan engagement platform, raised a $4 million seed round led by Game Changers Ventures with backing from Scooter Braun. The company pivoted to sell AI-parsed fan content…
Text-to-video AI has matured from experimental clips to hyper-realistic cinematic content, driven by a shift from U-Net to Diffusion Transformer (DiT) architectures. DiT offers linear scalability with compute and global …
RewindOS, a private, open-source alternative to Windows Recall and Rewind.ai, launches for Linux with fully local screen capture, OCR, and AI search. The MIT-licensed tool runs entirely on-device with no cloud, account, …
The University of Pennsylvania's Safe Autonomous Systems Lab (xLAB) and Built Robotics are partnering to develop physical AI for construction sites. Built will provide its dataset and a new data-collection robot to help …
Adobe released a broad Creative Cloud update across Lightroom, Lightroom Classic, Photoshop, Premiere Pro, After Effects, and Illustrator, adding AI features such as Assisted Culling, Photo to Video using Adobe Firefly a…
Researchers from KAIST introduced Video-based Optimal Transport Preference (VOTP), a method that uses optimal transport over Video Foundation Model embeddings to generate high-fidelity pseudo-labels from a small number o…
Echohive and Claude released 'A Survey of Monoliths', an artistic short film in HTML featuring live music and interactive controls, built with Three.js scenes. The film allows viewers to click, skip, pause, mute, and rep…
MoVerse achieves 8 FPS real-time 360° walkthrough video on a single NVIDIA RTX 4090, demonstrating that interactive generative video no longer requires a multi-GPU cluster. The system uses a causal autoregressive rendere…
Researchers introduced VigilFormer, a video anomaly detection framework combining deformable spatio-temporal attention with causal temporal modeling, achieving state-of-the-art AUC scores of 87.83% on UCF-Crime, 97.21% o…
Researchers propose Steady-Forcing, a memory and training framework for autoregressive video diffusion models that balances spatial persistence and motion continuity in long-horizon nature video generation. The method im…
Researchers introduced RAMS, a runtime controller that dynamically switches between YOLOv8 model tiers on embedded devices to balance latency and detection quality under resource pressure. On Jetson Orin TensorRT under h…
Researchers introduced AgentViSS, a benchmark for evaluating visual social intelligence in multimodal agents, featuring 240 scenarios and four role-level tasks. Tests on seven multimodal large language models revealed th…
Researchers propose Visual-Seeker, a multimodal deep search agent that actively reasons over visual details to perform multi-hop, cross-modal search. The agent achieves state-of-the-art results on five benchmarks, surpas…
Researchers introduced an LLM-driven framework for retrieving remote sensing data from cloud-based geospatial catalogues using natural language queries. The system uses three agents—Guardrail, General-QA, and Recommender…
Researchers introduced a foundation model-driven framework for cross-modal representation alignment between CT imaging and longitudinal EHR data, improving time-to-event prediction across tasks and institutions. Fusion s…
Researchers propose sub-semantic image segmentation, a new category that partitions images into stable appearance patterns described by language rather than naming whole objects. They introduce DETECTURE, a system that o…
Researchers developed a high-throughput 3D behavioral phenotyping framework integrating deep learning object detection with binocular stereo vision for real-time monitoring of juvenile tilapia. The system automates body …