Skip to main content
← Software and research artifacts

Research Radar · Experimental implementation

WebGenAI

Classifying coarse web-layout regions in synthetic mockups with a patch-based neural network.

Open repository ↗ Documentation ↗

What it investigates

WebGenAI investigates whether a patch-based model can recover coarse layout regions from synthetic mockup inputs.

Implementation record

The repository generates labeled mockups, trains a patch-based CNN/MLP classifier, and reconstructs header, navigation, card, and footer bounding boxes from patch predictions. It also contains an experimental ANE hook and an estimated comparison path; neither establishes ANE execution.

Public record

This build was recorded as a synthesis, with no single public source paper attached to it. The repository and its documentation are the public record for inspecting its approach. This page does not infer a source result or claim a reproduction.

Scope and limitations

This synthesis build has no single recorded public source paper. It produces coarse regions from synthetic labels, not optimized web components or production design-to-code output, and its ANE timing is estimated rather than hardware-measured.