Computer Vision news and analysis on Web Pulse: 1697 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Researchers introduced DMV-Bench, the first interactive benchmark for multimodal-agent visual memory, testing agents' ability to recall visual cues from product images in a simulated e-commerce environment. Their propose…
Researchers introduced TruEye, a model that detects and localizes AI-generated or manipulated humans in images with fine-grained classification across five compositional categories, including real humans composited into …
Researchers have developed OverFlowLight, a real-time framework that uses multi-modal sensing from cameras and radars to detect and prevent queue overflow at urban intersections. Deployed across 43 intersections in three…
Researchers propose LaViD, a framework that transfers semantic knowledge from a large language model to a vision-only student model using multiple-choice questions, outperforming existing distillation methods on fine-gra…
Researchers introduced P3Sim, a physical world modeling system that simulates future scene states from partial observations and incomplete 3D transformation signals. The system combines a learned world model, geometric c…
Researchers propose a new RANSAC scoring method that eliminates the need for a user-supplied inlier scale parameter by marginalizing it analytically under a conjugate Inverse-Gamma prior. The method achieves state-of-the…
Researchers propose vMFProto, a distributional part-prototype framework that models each class as a mixture of von Mises-Fisher components on the hypersphere for interpretable classification. The method achieves state-of…
Researchers introduced SelectAnyTree, a promptable instance segmentation model for 3D forest LiDAR point clouds that delineates individual trees from a few clicks. The model achieves 78.2% Intersection over Union (IoU) f…
Researchers introduced Tessellating the Earth (TTE), a learnable location encoder using Spherical Voronoi partitions that allocates representational capacity to discriminative areas. TTE sets a new state of the art in ge…
Researchers introduced NormAct, a benchmark evaluating whether multimodal large language models comply with hidden social norms during embodied planning. Tests on GPT-5.4, Claude Opus 4.7, and Gemini 3 Pro showed models …
Researchers propose a framework to expand human motion generation by using large-scale synthetic motion data and a redesigned VQ-VAE tokenizer, overcoming the limited diversity of motion capture datasets. The approach im…
JD.com launched Oxygen AIIC, an industrial-scale platform using LLMs/VLMs to manage item knowledge for tens of billions of SKUs. The platform achieves 94.2% precision and 82.8% recall in knowledge production, processes h…
Researchers propose CoIn, a novel framework for 3D scene inpainting that bridges 2D inpainting models and Gaussian Splatting through a multi-stage consistency pipeline. The method handles both object removal and insertio…
Researchers propose Transformation-Aware Decoupling (TAD), a framework for 3D scene graph generation that improves robustness to viewpoint changes by decoupling relation reasoning into stable and directional components. …
Researchers developed a distribution-based deep multiple instance learning method for tumor proportion scoring in non-small cell lung cancer, using only slide-level labels to predict zero-inflated beta parameters. The ap…
Researchers introduced PCB-Bench, the first comprehensive benchmark for evaluating large language models on printed circuit board placement and routing tasks. The benchmark includes over 3,700 questions across three task…
A new report by Mijente, Just Futures Law and the Surveillance Resistance Lab alleges that U.S. immigration authorities have expanded their use of surveillance technologies from private firms like Palantir and Anduril, r…
Vortx-AI open-sourced Emem, a signed satellite image service that enables AI agents to leverage physical-world data in a repetitive, cite-able manner, positioning it as a real-world intelligence layer similar to HTTPS.
A developer used a coding agent to combine Strava activity data with photos of gym workout boards to generate a half-year training recap infographic. The agent extracted exercise details from images and classified activi…
Bytedance Seedance V1.5 Pro Image To Video defeated Marey Realism V1.5 in a head-to-head test, scoring 17.2 to 13.4, because it better obeyed the prompt's instructions for specific physical events and scene geography. Th…