A developer built a cover-only book recommender that fuses CLIP vector embeddings with GLiNER-extracted entities via Reciprocal Rank Fusion, queried against the Hardcover API, and uses a two-tower neu…
Bongard problems, visual puzzles that challenge AI to identify latent properties distinguishing two sets of images, remain a benchmark for visual reasoning, as highlighted in a new essay revisiting Do…
PROVE (Prompt Recovery with Verified Evidence), a training-free, black-box prompt inversion attack introduced in arXiv:2608.13671v1, reconstructs prompts by composing verifiable scene descriptions and…
ReClaim, a computer-vision-powered lost-and-found system for campus communities, uses CLIP embeddings and cosine similarity to match lost items with found items. The system, built with FastAPI and PyT…
A new open-world perception system for driving scenes safely handles unknown road objects by using a hierarchical taxonomy, abstraction fallback, and segmentation cross-validation, as described in a r…
Harness, a screen-aware AI assistant, runs its entire perception stack in the browser to keep costs at zero for high-frequency tasks. The system uses on-device models like CLIP, PaddleOCR, and a quant…
A developer built a custom multimodal Retrieval-Augmented Generation (RAG) system that reads and understands thousands of PDFs using open-source AI, extracting text and images, chunking content, and e…
An AWS developer built a zero-shot image classifier using a CLIP model inside a Lambda container to distinguish AWS Builder Cards from other images without any training data or GPU. The model, which e…
A developer built Visual Forensics Radar, an 'Ensemble of Experts' system that combines Error Level Analysis, Zero-Shot CLIP classification, and a Vision-Language Model (Qwen2-VL) to detect hybrid dee…
Researchers introduced EB-CaP, a test-time adaptation method for video facial expression recognition that personalizes class prototypes per target video using an energy-based model and CLIP embeddings…
Meta AI's DINOv2 (2023) scales the original DINO self-supervised vision model into a Vision Foundation Model that matches or beats text-guided models like CLIP on dense pixel-level tasks, introducing …
A new dissertation from an unnamed researcher introduces bio-inspired active vision systems that use foveal-peripheral sampling and learned saccades to improve efficiency and robustness in deep neural…
A new tutorial from StarTrail-org introduces PixelRAG, a pixel-native retrieval-augmented generation pipeline that renders web pages and PDFs as images, divides them into overlapping tiles, and genera…
OpenAI researchers demonstrated in 2021 that CLIP, a vision language model, misclassifies a Granny Smith apple as an iPod when a paper label reading 'iPod' is attached, revealing that grounding in AI …
Researchers introduce TraceCLIP, a training-free framework that recovers latent patch-level semantic evidence from CLIP's CLS attention output, achieving gains of 1.3 to 4.5 points in average mIoU ove…
Researchers introduce Diff-ID, a diffusion-based framework that enforces identity consistency in facial image generation while delivering photorealistic quality, using a custom 210K image dataset and …
MorphUNet, a diffusion-based face morphing framework introduced in a new arXiv preprint (2607.25092v1), achieves the best Morphing Attack Potential (MAP) of 0.919 on FEI and 0.886 on FRLL when at leas…
Researchers propose TextSLIP, a medical vision-language pretraining framework that augments CLIP with intra-modal text contrastive learning to improve fine-grained semantic supervision for radiology r…
MIT has implemented an AI video surveillance system that converts raw video feeds into structured, searchable metadata using large language models. The system captures streams via RTSP or WebRTC, extr…
A new study from arXiv (2607.18695v1) finds that descriptors generated by large language models (LLMs) for zero-shot classification carry little visual evidence, collapsing ImageNet accuracy from 59.5…