Can a strong general-purpose vision model distinguish real from AI-generated images using only a simple decision rule on frozen representations? Our latest results, to be presented at #ECCV2026, show it can.
Blog: https://t.co/EvTnCIqkB5
Paper: https://t.co/NzvieckZlh
New image generators keep appearing. A trained AI-generated image detector that works well on familiar images can fail when the generator, prompt, style, or image domain changes. Our benchmark study introduced Percept-Lens, a common evaluation framework for these shifts, and showed how sharply released AI-generated image detectors can degrade beyond familiar data.
That led us to a more basic question. When a detector fails, has its underlying vision model lost the distinction between real and AI-generated images, or is its decision rule failing to recover it?
In our upcoming ECCV paper, we built a new detector by keeping a general-purpose vision model frozen and fitting a simple Gaussian decision rule to its representations. The method models how labeled real and AI-generated images are arranged in the vision model’s feature space, then classifies a new image by the group it most closely resembles.
On the same…
Detector generalisation failures come from the decision rule, not from lost representation. Fitting a simple Gaussian classifier on a frozen vision model's features outperforms purpose-trained detectors across unfamiliar generators and domains.
Read the full post on X ↗
