AI scans 450+ surveillance cameras to find Wally
AI scans 450+ surveillance cameras to find Wally
An open-source AI model reviews footage from unprotected cameras worldwide to find people who resemble Wally from Where's Wally?.
AI scans 450+ surveillance cameras to find Wally
Belgian artist Dries Depoorter created an unusual art project: a neural network reviews recordings from unprotected cameras and looks for people resembling the character from Where's Wally?.
The search uses the open model Qwen3-VL from Alibaba, which runs locally on three gaming graphics cards. AI analyzes each frame and looks for the distinctive red-and-white striped clothing and a hat with a pom-pom.
So far, the system has checked 1.68 million frames and found 13 matches in different countries, including the United States, Japan, and Peru. The artist manually verifies each find.
The game of hide-and-seek has reached a new level. More details are available on WallyFinder.
Why it matters
AnalysisThe project shows how open multimodal models can turn large video streams into a research tool, while also highlighting privacy and security issues around unprotected cameras.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
Jev and System One AI: A Complete Practical Guide to the Decision Model from TypeSafe
A deep engineering breakdown of the flagship Jev model: System One architecture versus LLMs, Choice/Score/Noul primitives, request parallelism, confidence mathematics, video use cases, code, and an overview of local open-weight alternatives (Laya, GLiNER2.5, CLM-8B).
JSON — Structured Data for Programs, APIs, and AI Agents
A complete hands-on guide to JSON: syntax specifications, serialization via parse and stringify, REST API consumption, JSON Schema validation, and structured outputs for LLMs.
Multimodal Embeddings (CLIP)
This technology designs various media types (text, images, audio) into a unified shared multidimensional vector space using architectures like OpenAI CLIP or Google SigLIP. It enables searching for images using textual descriptions or finding similar music based on pictures.
Hugging Face Hub
The leading global platform and open repository for the AI community ('GitHub for Neural Networks'). It hosts hundreds of thousands of open models, datasets, and free web demos (Spaces), supporting all major AI frameworks.