Picture for Ismail Elezi

Ismail Elezi

VeLoRA: Memory Efficient Training using Rank-1 Sub-Token Projections

May 28, 2024
Viaarxiv icon

Deep Active Learning: A Reality Check

Mar 21, 2024
Viaarxiv icon

$V_kD:$ Improving Knowledge Distillation using Orthogonal Projections

Add code
Mar 10, 2024
Figure 1 for $V_kD:$ Improving Knowledge Distillation using Orthogonal Projections
Figure 2 for $V_kD:$ Improving Knowledge Distillation using Orthogonal Projections
Figure 3 for $V_kD:$ Improving Knowledge Distillation using Orthogonal Projections
Figure 4 for $V_kD:$ Improving Knowledge Distillation using Orthogonal Projections
Viaarxiv icon

G3DR: Generative 3D Reconstruction in ImageNet

Add code
Mar 08, 2024
Figure 1 for G3DR: Generative 3D Reconstruction in ImageNet
Figure 2 for G3DR: Generative 3D Reconstruction in ImageNet
Figure 3 for G3DR: Generative 3D Reconstruction in ImageNet
Figure 4 for G3DR: Generative 3D Reconstruction in ImageNet
Viaarxiv icon

Three Heads Are Better Than One: Complementary Experts for Long-Tailed Semi-supervised Learning

Add code
Dec 25, 2023
Viaarxiv icon

Data-Driven but Privacy-Conscious: Pedestrian Dataset De-identification via Full-Body Person Synthesis

Jun 22, 2023
Figure 1 for Data-Driven but Privacy-Conscious: Pedestrian Dataset De-identification via Full-Body Person Synthesis
Figure 2 for Data-Driven but Privacy-Conscious: Pedestrian Dataset De-identification via Full-Body Person Synthesis
Figure 3 for Data-Driven but Privacy-Conscious: Pedestrian Dataset De-identification via Full-Body Person Synthesis
Figure 4 for Data-Driven but Privacy-Conscious: Pedestrian Dataset De-identification via Full-Body Person Synthesis
Viaarxiv icon

Learning to Discover and Detect Objects

Add code
Oct 19, 2022
Figure 1 for Learning to Discover and Detect Objects
Figure 2 for Learning to Discover and Detect Objects
Figure 3 for Learning to Discover and Detect Objects
Figure 4 for Learning to Discover and Detect Objects
Viaarxiv icon

The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes

Add code
Oct 13, 2022
Figure 1 for The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes
Figure 2 for The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes
Figure 3 for The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes
Figure 4 for The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes
Viaarxiv icon

Simple Cues Lead to a Strong Multi-Object Tracker

Add code
Jun 13, 2022
Figure 1 for Simple Cues Lead to a Strong Multi-Object Tracker
Figure 2 for Simple Cues Lead to a Strong Multi-Object Tracker
Figure 3 for Simple Cues Lead to a Strong Multi-Object Tracker
Figure 4 for Simple Cues Lead to a Strong Multi-Object Tracker
Viaarxiv icon

The Group Loss++: A deeper look into group loss for deep metric learning

Apr 04, 2022
Figure 1 for The Group Loss++: A deeper look into group loss for deep metric learning
Figure 2 for The Group Loss++: A deeper look into group loss for deep metric learning
Figure 3 for The Group Loss++: A deeper look into group loss for deep metric learning
Figure 4 for The Group Loss++: A deeper look into group loss for deep metric learning
Viaarxiv icon