Computer Vision news and analysis on Web Pulse: 1728 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
A developer with over two decades of experience in technology describes how artificial intelligence has become an invisible infrastructure in daily life, citing practical use cases such as LLM-based productivity assistan…
AMD launched a sweepstakes giving away six Radeon GPUs, including the flagship RX 9070 XT and RX 7900 XTX, to celebrate FSR 4.1 support for the RX 7000 series. The giveaway is open to US, UK, Ireland, and Canada (excludi…
A new PyTorch-based optimization method reconstructs high-resolution Sentinel-2 reflectance scenes from approximately 30 low-resolution observations by exploiting sub-pixel jitter between repeat passes, using Gaussian sp…
Germany's Rogon Technologies built Cuju, an AI scouting app that uses standardized video drills to evaluate young soccer players in Brazil, replacing traditional scouting methods. The app has 160,000 downloads by mid-202…
ByteDance released Seedance 2.5, an AI video generator supporting text-to-video and image-to-video creation for marketing content. The tool offers brand consistency, multi-platform formats, and production-ready exports f…
Biostar announced the EdgeComp MS-NAT5000, a compact edge AI computing system integrating the NVIDIA Jetson Thor T5000 module for generative AI, visual intelligence, and real-time sensor processing in industrial and auto…
Photoroom reduced inference latency of its convolutional segmentation backbone by 22% on A100 GPUs by switching to PyTorch's channels-last memory format. The change required only four lines of code and did not affect acc…
NC AI, a Korean AI company spun off from NCSoft, released VARCO 3D 2.0, a next-generation 3D asset generation model that achieved top benchmark scores, including a 40.8% improvement on the Uni3D metric over its predecess…
An Associated Press feature profiles a wave of startups and researchers shifting focus from language models to 'world models' that learn physical space and interaction. Computer scientist Louis Castricato left Brown Univ…
East Kazakhstan is deploying an AI-powered early wildfire detection system across nine forestry enterprises, installing 90 antenna-mast complexes with cameras and automated image analysis to automatically detect smoke an…
A study benchmarking six RAG architectures on a financial PDF reveals that Vision RAG achieves the highest answer correctness (0.823) but low faithfulness (0.538), while Graph RAG is the most faithful (0.902) yet least c…
Darktable 5.6 introduces its first AI features, including an AI object mask and a neural restore module for denoising and upscaling, all running locally on the user's CPU or GPU. The open-source photo editor emphasizes o…
Researchers propose Forensic Knowledge Graphs (FKGs) to unify forensic evidence extraction, structured reasoning, and interpretable explanations for image authentication. The framework outperforms existing forensic detec…
Researchers propose ObsGraph, a hierarchical scene graph that unifies scene representation, retrieval, and exploration for embodied reasoning. The system organizes visual evidence into room-view-object layers and uses co…
Researchers propose TheProfessor, a multi-teacher prompt distillation method for vision-language models that ensembles a domain-finetuned teacher and a zero-shot teacher. On four datasets, confidence-weighted ensembling …
Researchers introduced REALM, the first unified red-teaming benchmark for physical-world vision-language models (VLMs), integrating 12 attack methods, 3 defenses, and 13 VLMs under a black-box threat model. The benchmark…
Researchers introduced HANCLIP, a family of vision-language models that restructure the embedding space to handle negation by combining hyperbolic and angular objectives. Trained on 20,000 image-text quadruplets, HANCLIP…
Researchers propose a frequency-guided feature representation learner for small object detection that shifts from spatial to spectral processing. The method, instantiated via lightweight plug-and-play modules, achieves c…
Researchers propose Prompt-Vision Token Activation Map (PV-TAM) to improve vision-language consistency in large vision-language models by using prompt-side semantics and filtering systematic bias from modality boundary m…
Researchers propose a training-free IBA framework for Knowledge-Based VQA that decouples entity identification from evidence re-ranking, outperforming fine-tuned baselines on Encyclopedic-VQA and InfoSeek while reducing …