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.
A developer built ClearPix, a static, backend-free watermark remover that runs ONNX Runtime Web inference entirely in the browser via WebGPU with a WASM fallback, so user images and videos are never uploaded. The enginee…
Researchers at the University of Alabama trained a model only on video footage captured prior to violent events — combining body motion, facial disposition and audio — to test whether trainable 'pre-violent' characterist…
A developer on the Mendly team built a computer-vision pipeline for stroke rehabilitation that converts MediaPipe Pose landmarks into joint angles, range-of-motion measurements, and valid-repetition counts for an adaptiv…
Anaconda now lists 77 curated, vetted open-source AI models on anaconda.org, up from its prior text-generation catalog after adding eight models, five of them vision-language models for document understanding and image a…
Roboflow launched Roboflow Labs, a new applied research division focused on teaching models to see, reason, and act in three dimensions across industries, the company announced. Roboflow Labs will publish research and tr…
Meituan released LongCat-2.5-Preview on its API platform on September 25, adding image understanding to the company's AI model lineup. According to the model's documentation, LongCat-2.5-Preview can answer questions abou…
A developer built an image analyzer that chains Google Cloud Vision API with a FastAPI service hosted on Cloud Run to automatically label, tag, and moderate user-uploaded images. The serverless architecture scales to zer…
A forum thread on fine-tuning CLIP-based NSFW detection models surfaced practical guidance from practitioners, including a suggested starting point of batch size 16 and learning rate 5e-5 drawn from a ViT tuning document…
NVIDIA researchers, with MIT and the University of Oxford, introduced Physis-Lang, a self-evolving physical-language framework that ranks first on the Physics-IQ Verified leaderboard snapshot dated September 29, 2026 wit…
A writer built a YOLO-based object detector in about 90 minutes to flag "Excited to announce I made a YOLO project" posts on LinkedIn, after collecting 200 screenshots of their feed with a scrot/xdotool script and annota…
A new arXiv paper (2609.36172v1) reports that VGG16 achieved 89.55% validation accuracy in recognizing topological relationships between objects in images, outperforming traditional models including Naive Bayes, KNN, Ran…
PreviewDiff, a training-free test-time search method described in arXiv paper 2609.36199v1, turns diffusion sampling from scalar search into a multimodal critic-guided search over intermediate latents, decoding partial p…
A new arXiv paper (2609.37175v1) presents a general-purpose vision-language solver that augments textual instance descriptions with input-derived visual representations for end-to-end combinatorial optimization, trained …
HeadGuard, a composable head-protection method for low-bit vision-language model KV-cache quantization, recovers a substantial fraction of lost accuracy by keeping roughly 1/8 of physical KV heads in bfloat16, according …
A new arXiv paper (2609.35910v1) introduces DEEP (Decision Evaluation for Escalated Perception), a benchmark that scores pre-escalation compute allocators against a budget-constrained decision oracle, finding that 34–54%…
A team of researchers has developed an Optical Character Recognition (OCR) model that combines a deep learning model with image preprocessing techniques to extract information from Korean-language invoices, achieving an …
Researchers introduced SABLE (Sparse-view Animal Behavior Latent Embeddings), a self-supervised framework that reconstructs 3D animal behavior from two-view video without ground-truth 3D labels, according to the arXiv pa…
Researchers introduced Persistence Forcing (PerF), a pixel-space diffusion Transformer method that assigns different feature groups distinct refinement budgets across depth, producing what the authors call persistent and…
CoDimRecon, an agentic framework detailed in arXiv paper 2609.36024v1, reconstructs editable, simulation-ready 3D scenes containing rigid, articulated, and deformable objects from multi-view RGB observations. The framewo…
A new arXiv paper (2609.36101v1) reports that a single dominant direction captures 94.4-99.9% of the squared norm of the image-text mean separation in CLIP and SigLIP vision-language encoders, explaining why modifying th…