Ali K. Thabet

MAIN: Multi-Attention Instance Network for Video Segmentation

Instance-level video segmentation requires a solid integration of spatial and temporal information. However, current methods rely mostly on domain-specific information (online learning) to produce...

Video Self-Stitching Graph Network for Temporal Action Localization

Learning to Cut by Watching Movies

DeepGCNs: Can GCNs Go As Deep As CNNs?

MAAS: Multi-modal Assignation for Active Speaker Detection

Re-ReND: Real-time Rendering of NeRFs across Devices

Reconfigurable Snapshot HDR Imaging Using Coded Masks and Inception Network

Rethinking Clustering for Robustness

BoDiffusion: Diffusing Sparse Observations for Full-Body Human Motion Synthesis

Enhancing Adversarial Robustness via Test-time Transformation Ensembling

Bespoke Solvers for Generative Flow Models

Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment

ASSANet: An Anisotropic Separable Set Abstraction for Efficient Point Cloud Representation Learning

VisCo Grids: Surface Reconstruction with Viscosity and Coarea Grids

Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model

SGAS: Sequential Greedy Architecture Search

Autoregressive Distillation of Diffusion Transformers

Towards Characterizing the Semantic Robustness of Face Recognition