Federico Tombari

Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video Generator

AnyUp: Universal Feature Upsampling

LaRI: Layered Ray Intersections for Single-view 3D Geometric Reasoning

RefAM: Attention Magnets for Zero-Shot Referral Segmentation

Adapted-Language ViT: Empowering Self-Supervised Vision Transformers with LLMs

DRGSplat: Depth-Regularized 3D Gaussian Splatting

SG-Tailor: Inter-Object Commonsense Relationship Reasoning for Scene Graph Manipulation

Epipolar Geometry Improves Video Generation Models

3D-LATTE: Latent Space 3D Editing from Textual Instructions

CoReVe: Mitigating Object Hallucinations in Large Vision-Language Models via Chain-of-Region Verification

CausNVS: Autoregressive Multi-view Diffusion for Flexible 3D Novel View Synthesis

SG2Loc: Sequential Visual Localization on 3D Scene Graphs

CubeDiff: Repurposing Diffusion-Based Image Models for Panorama Generation

TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters

LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering

Learning Neural Exposure Fields for View Synthesis

Video Perception Models for 3D Scene Synthesis

Gatekeeper: Improving Model Cascades Through Confidence Tuning

Object-X: Learning to Reconstruct Multi-Modal 3D Object Representations

Epipolar Geometry Improves Video Generation Models

UIP2P: Unsupervised Instruction-Based Image Editing via Cycle Edit Consistency

CLoRA: A Contrastive Approach to Compose Multiple LoRA Models

Splat-SLAM: Globally Optimized RGB-only SLAM with 3D Gaussians

How Good is my Video LMM? Complex Video Reasoning and Robustness Evaluation Suite for Video-LMMs

ESCAPE: Equivariant Shape Completion via Anchor Point Encoding