Computer Vision news and analysis on Web Pulse: 4325 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Gujarat is building a Rs 150-crore AI-powered Command and Control Centre in Gandhinagar to track Asiatic lions, sloth bears, black bucks, and wild asses across protected habitats, Forest and Environment Minister Arjun Mo…
A new browser-based tool called Vajba lets users check a photo for signs of editing, AI generation, hidden metadata, and compression artifacts without uploading the file to a server. The tool analyzes JPEG, PNG, WebP, HE…
Apple has introduced Apple Reference Image, a sensor-level cryptographic signing feature debuting on the iPhone 18 Pro that signs every pixel the camera captures and stores an "unalterable reference image" alongside the …
A developer's analysis of the aiappdex.com model catalog highlights three non-text AI models seeing heavy production use: Amazon's Chronos-2 for zero-shot time-series forecasting (25.7M downloads), OpenAI's CLIP-vit-base…
Clearview AI has been quietly testing a prototype tool called InquiryIQ that uses xAI's Grok model to surface associates, social media accounts, and other personal details about people identified through its facial recog…
A survey examines inference-efficiency mechanisms in video and audiovisual large language models (VideoLLMs), which pair video representations with pretrained large language models and condition generation on a textual p…
A developer documented how browser-based canvas editors produce faint vertical seams in exported images even when the on-screen preview looks flawless. The artifacts stem from high-resolution filters being processed in h…
Hanwha Systems unveiled its Space AI Analytics Solution at MSPO 2026 in Kielce, Poland, the Korean defense company's first overseas introduction of the technology, which analyzes satellite imagery in real time. The syste…
Researchers introduced DF26, a new benchmark for detecting AI-generated videos made with recent text-to-video and image-to-video models. The benchmark covers single-person public-speaking scenarios, including direct-to-c…
Researchers posted arXiv:2609.07137v1, a paper introducing Flow3D-OPD, a two-stage post-training framework that applies multi-teacher on-policy distillation to image-to-3D generation built on flow-matching diffusion Tran…
Researchers introduced Capsule Lens, a framework that fits a simple geometric "capsule" defined by interpretable parameters in closed form to each concept's region in a machine learning model's internal representations, …
A new arXiv paper (2609.09528v1) introduces MotionBlind, a contrastive benchmark showing that Video-LLMs cannot reliably judge physically grounded motion, with open models sitting near the 6.25% chance floor and scale pr…
Researchers introduced VANTAGE-Bench, a benchmark measuring an "Infrastructure AI Gap" in Vision-Language Models (VLMs) that rely on fixed cameras for open-loop tasks such as safety monitoring and operational logging. Ev…
A new arXiv paper (2609.05448v1) formulates structured post-training pruning of vision and language transformers as a damage-aware multi-armed bandit problem under a fixed candidate-evaluation budget, masking attention h…
Researchers released Video-MOPD-8B, an open-weight video understanding model trained via Multi-Teacher On-Policy Distillation (MOPD), which the paper reports achieves state-of-the-art performance among models of comparab…
A study of vision-language model reliability found that a frozen Qwen2.5-VL-3B-Instruct model showed an evidence monotonicity violation rate (EMVR) of 0.436, with 92.0% of 176 accepted GQA-derived trajectories containing…
A new arXiv paper (2609.09462v1) reports that factorizing the dense CoOp prompt matrix P ∈ R^(m×d) as P = BA cuts trainable prompt parameters from md to r(m+d), and to rd once the token-side factor B is fixed, matching o…
A new deep, global Structure-from-Motion framework learns camera poses from noisy view-graphs using a permutation-equivariant, edge-conditioned graph neural network that outputs globally consistent camera extrinsics with…
Records obtained by Inadvertent show the Northern California Regional Intelligence Center, a law enforcement fusion center, conducted more than 27,000 searches accessing Santa Ana's Flock Safety license plate reader netw…
ML Systems is proposing a "Collective Ontology" — a shared grammar of typed codes and directed-acyclic-graph edges meant to let assessors, vision models, estimators, and eventually robotic crews describe a single roof as…