8 papers
ReFP-AD: Rectified Flow Preconditioning for Energy-Based Anomaly Detection
Camile Lendering, Erkut Akdag, JoaquÃn Figueira +1
Unified anomaly detection requires modeling highly heterogeneous normal data without access to anomalous samples. While foundation models like DINOv2 provide rich token representat…
SubspaceAD: Training-Free Few-Shot Anomaly Detection via Subspace Modeling
Camile Lendering, Erkut Akdag, Egor Bondarev
Detecting visual anomalies in industrial inspection often requires training with only a few normal images per category. Recent few-shot methods achieve strong results employing fou…
Boxes2Pixels: Learning Defect Segmentation from Noisy SAM Masks
Camile Lendering, Erkut Akdag, Egor Bondarev
Accurate defect segmentation is critical for industrial inspection, yet dense pixel-level annotations are rarely available. A common workaround is to convert inexpensive bounding b…
F3G-Avatar : Face Focused Full-body Gaussian Avatar
Willem Menu, Erkut Akdag, Pedro Quesado +2
Existing full-body Gaussian avatar methods primarily optimize global reconstruction quality and often fail to preserve fine-grained facial geometry and expression details. This cha…
R3PM-Net: Real-time, Robust, Real-world Point Matching Network
Yasaman Kashefbahrami, Erkut Akdag, Panagiotis Meletis +3
Accurate Point Cloud Registration (PCR) is an important task in 3D data processing, involving the estimation of a rigid transformation between two point clouds. While deep-learning…
LiveStre4m: Feed-Forward Live Streaming of Novel Views from Unposed Multi-View Video
Pedro Quesado, Erkut Akdag, Yasaman Kashefbahrami +2
Live-streaming Novel View Synthesis (NVS) from unposed multi-view video remains an open challenge in a wide range of applications. Existing methods for dynamic scene representation…