Bernard Ghanem

27

Papers

255

Total Citations

Papers (27)

GES : Generalized Exponential Splatting for Efficient Radiance Field Rendering

End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames

BOLT: Boost Large Vision-Language Model Without Training for Long-form Video Understanding

Adaptive Guidance: Training-free Acceleration of Conditional Diffusion Models

Efficient Image Pre-Training with Siamese Cropped Masked Autoencoders

Generalizability of Adversarial Robustness Under Distribution Shifts

Privacy-Preserving Optics for Enhancing Protection in Face De-Identification

Towards Automated Movie Trailer Generation

SynFER: Towards Boosting Facial Expression Recognition with Synthetic Data

DATENeRF: Depth-Aware Text-based Editing of NeRFs

SimCS: Simulation for Domain Incremental Online Continual Segmentation

SMILE: Infusing Spatial and Motion Semantics in Masked Video Learning

Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives

Evaluation of Test-Time Adaptation Under Computational Time Constraints

3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes

Towards Interpretable Deep Local Learning with Successive Gradient Reconciliation

Enhancing Online Continual Learning with Plug-and-Play State Space Model and Class-Conditional Mixture of Discretization

Diffusion-Based Imaginative Coordination for Bimanual Manipulation

HAMSt3R: Human-Aware Multi-view Stereo 3D Reconstruction

MatchDiffusion: Training-free Generation of Match-Cuts

4D-Bench: Benchmarking Multi-modal Large Language Models for 4D Object Understanding

UnMix-NeRF: Spectral Unmixing Meets Neural Radiance Fields

ResidualViT for Efficient Temporally Dense Video Encoding

OmniResponse: Online Multimodal Conversational Response Generation in Dyadic Interactions

SPAD: Spatially Aware Multi-View Diffusers

Dr2Net: Dynamic Reversible Dual-Residual Networks for Memory-Efficient Finetuning

Tune-An-Ellipse: CLIP Has Potential to Find What You Want