Computer Vision news and analysis on Web Pulse: 4314 curated articles tracking the latest Computer Vision developments, tools, and research, updated continuously from vetted sources.
Xiaomi researchers released Xiaomi-OCR-0, a unified 0.8B-parameter vision-language model for document parsing and OCR-centric understanding, built on Qwen3.5-0.8B and trained on an approximately 170M-sample OCR-centric c…
A new arXiv paper (arXiv:2609.36219v1) introduces LeRF, a framework that trains Vision-Language Models to build and use explicit reference coordinate frames for perspective-taking reasoning, addressing the tendency of VL…
A new arXiv paper (2609.36348v1) reports that adding cross-view class-token alignment to diffusion transformer training improves ImageNet linear-probing accuracy by 9.4% with the class token and 10.1% with mean-pooled pa…
Researchers introduced Imagine3D-LLM, a multimodal large language model that decodes a small set of learnable summary tokens into a compact 3D Gaussian Splatting representation of a scene, supervised by a photometric rec…
Adversarial learning can serve as an effective post-training correction for pixel diffusion models, which generate RGB images directly without an autoencoder but systematically underrepresent fine-scale natural-image sta…
VideoLoop, a new method for long-form video understanding, counters "semantic thrashing" in multimodal agents by looping working memory instead of appending to it, according to the paper's authors. The approach targets t…
Mininglamp-AI released Mano-P, an open-source GUI agent that runs entirely on-device on Apple Silicon Macs, using a 4B vision-language model quantized to MLX 8-bit via the Cider SDK. The agent follows an observation-reas…
An empirical study examines why World Action Models (WAMs) generalize, comparing Explicit WAMs, which denoise future frames into clean video alongside every action chunk during inference, against Latent WAMs, which avoid…
A developer released Image Categorizer, an open-source tool that sorts a photo library into categories using local vision and text models served by Ollama, keeping all photos on the user's own network. The tool describes…
A developer built AniFlow, an open-source (MIT) self-hosted AI studio that converts manhwa and webtoon chapters into finished narrated recap videos. The pipeline chains YOLOv8 panel detection, speech-bubble detection and…
Software engineer Abdulrahman Haramain built Darb, an assistive navigation platform for the visually impaired that splits inference between a zero-latency on-device vision engine running MediaPipe at 30+ FPS and a cloud-…
Roboflow published a guide for building an AI visual-inspection agent that chains RF-DETR defect detection, object tracking, and Google's Gemini vision-language model inside Roboflow Workflows to inspect carton packages …
NVIDIA released VSS Blueprint 3.3, adding a Build Vision Agent skill (vss-build-vision-ai) and Adaptive Efficient Video Sampling (EVS) that cut VLM input tokens by 80% for a 60-minute summary and raise concurrent streams…
OpenAI launched the Decisions API, a constrained decision-making interface powered by its Luna model that supports visual inputs and is tuned for end-to-end decisions in less than a few hundred milliseconds, according to…
Simon Willison released Photo Scrubber, a browser-based tool that automatically detects and blurs faces in photographs and strips metadata, built with GPT-6 Astra. The tool runs locally using Google's MediaPipe C++ libra…
Researchers introduced SCOPD (Sparse-Context On-Policy Self-Distillation), a method for efficient vision-language models that addresses the performance degradation caused by training-free token pruning of long visual-tok…
OGAD (Off Grid AI Desktop) released version 0.0.54-beta.108, a free desktop app that lets designers and photographers run a local vision model to analyze their own reference photos for visible light, composition, texture…
A developer has released OGAD (Off Grid AI Desktop), a local gateway that exposes offline vision, transcription, image-generation and text-to-speech capabilities to apps through a single HTTP address on port 7878. The to…
A developer built a three-stage video deduplication worker in Python that combines SHA-256 byte hashing, a 64-bit perceptual hash from the videohash2 library checked via Hamming distance, and FFmpeg's MPEG-7 signature fi…
Off Grid AI's OGAM (mobile) and OGAD (desktop) apps let a phone send image questions to a vision model running on a local computer over a trusted private network, avoiding cloud AI providers. The setup uses the desktop's…