12 papers
When Rule Violations Are Rare: Chimera Training for Logical Anomaly Detection
Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz +2
Many practical anomalies are not merely rare inputs, but violations of semantic constraints: objects co-occur in structured ways, actions imply preconditions, and events satisfy te…
Testing the Test: Score-Direction Instability in Class-Split Anomaly Detection
Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz +2
Within-dataset class-split evaluation is widely used as a proxy for fully unconditional out-of-distribution anomaly detection. We show that this protocol can become ill-posed when…
High-Dimensional Latents Should Be Diagnosed Through Phase Structure
Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz +2
We study autoencoder and variational-autoencoder latent spaces through the lens of spin-glass theory. The paper has two components. First, we formalize a latent-space spin-glass di…
VAE with Hyperspherical Coordinates: Improving Anomaly Detection from Hypervolume-Compressed Latent Space
Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz +2
Variational autoencoders (VAE) encode data into lower-dimensional latent vectors before decoding those vectors back to data. Once trained, one can hope to detect out-of-distributio…
First Shape, Then Meaning: Efficient Geometry and Semantics Learning for Indoor Reconstruction
Remi Chierchia, Léo Lebrat, David Ahmedt-Aristizabal +3
Neural Surface Reconstruction has become a standard methodology for indoor 3D reconstruction, with Signed Distance Functions (SDFs) proving particularly effective for representing…
In Depth We Trust: Reliable Monocular Depth Supervision for Gaussian Splatting
Wenhui Xiao, Ethan Goan, Rodrigo Santa Cruz +4
Using accurate depth priors in 3D Gaussian Splatting helps mitigate artifacts caused by sparse training data and textureless surfaces. However, acquiring accurate depth maps requir…