Lfr Eccv2026

Summary

Last-Layer-Centric Feature Recombination: Unleashing 3D Geometric Knowledge in DINOv3 for Monocular Depth Estimation. Explainability-Aware Frustum Attack: Exposing Structural Vulnerabilities in LiDAR-Based 3D Object Detectors. Title: ReconSplat: Generalizable 3D Scene Reconstruction Beyond Observed Views Authors: Giuseppe Stracquadanio, Kevin Raj ... Global Pose Control for Generative View Synthesis in Normalized Object Coordinate Space. Functional maps represent a correspondence between two 3D shapes as a linear matrix acting between their truncated spectral ... LaVPR: Benchmarking Language and Vision for Place Recognition | Our work, “Scaling Multi-Reference Image Generation with Dynamic Reward Optimization,” has been accepted to Beyond Alignment: A Generative Matching Paradigm for Zero-Shot Skeleton-based Action Recognition | This video presents the paper "FLEG: Feed-Forward Language Embedded Gaussian Splatting from Any Views via Compact ... Project page: superflex3d.github.io Authors: Gabriel Tavernini, Elisabetta Fedele, Tiago Novello, Leonidas Guibas, Marc ...