Joseph Tighe

papers

977

total citations

papers (20)

VidTr: Video Transformer Without Convolutions

ICCV 2021arXiv

220

citations

SiamMOT: Siamese Multi-Object Tracking

CVPR 2021arXiv

167

citations

Rethinking Zero-Shot Video Classification: End-to-End Training for Realistic Applications

CVPR 2020arXiv

138

citations

Combining Detection and Tracking for Human Pose Estimation in Videos

CVPR 2020arXiv

126

citations

TubeR: Tubelet Transformer for Video Action Detection

CVPR 2022arXiv

citations

What To Look at and Where: Semantic and Spatial Refined Transformer for Detecting Human-Object Interactions

CVPR 2022arXiv

citations

Joseph Tighe

papers (20)

VidTr: Video Transformer Without Convolutions

SiamMOT: Siamese Multi-Object Tracking

Rethinking Zero-Shot Video Classification: End-to-End Training for Realistic Applications

Combining Detection and Tracking for Human Pose Estimation in Videos

TubeR: Tubelet Transformer for Video Action Detection

What To Look at and Where: Semantic and Spatial Refined Transformer for Detecting Human-Object Interactions

Directional Temporal Modeling for Action Recognition

Hierarchical Self-Supervised Representation Learning for Movie Understanding

ScaleDet: A Scalable Multi-Dataset Object Detector

Large Scale Real-World Multi-person Tracking

MaCLR: Motion-Aware Contrastive Learning of Representations for Videos

Selective Feature Compression for Efficient Activity Recognition Inference

Threshold-Consistent Margin Loss for Open-World Deep Metric Learning

Single View Physical Distance Estimation Using Human Pose

An In-depth Study of Stochastic Backpropagation

SkeleTR: Towards Skeleton-based Action Recognition in the Wild

Exploiting weakly supervised visual patterns to learn from partial annotations

Id-Free Person Similarity Learning

PSS: Progressive Sample Selection for Open-World Visual Representation Learning

Learning for Transductive Threshold Calibration in Open-World Recognition

papers (20)

VidTr: Video Transformer Without Convolutions

SiamMOT: Siamese Multi-Object Tracking

Rethinking Zero-Shot Video Classification: End-to-End Training for Realistic Applications

Combining Detection and Tracking for Human Pose Estimation in Videos

TubeR: Tubelet Transformer for Video Action Detection

What To Look at and Where: Semantic and Spatial Refined Transformer for Detecting Human-Object Interactions

Directional Temporal Modeling for Action Recognition

Hierarchical Self-Supervised Representation Learning for Movie Understanding

ScaleDet: A Scalable Multi-Dataset Object Detector

Large Scale Real-World Multi-person Tracking

MaCLR: Motion-Aware Contrastive Learning of Representations for Videos

Selective Feature Compression for Efficient Activity Recognition Inference

Threshold-Consistent Margin Loss for Open-World Deep Metric Learning

Single View Physical Distance Estimation Using Human Pose

An In-depth Study of Stochastic Backpropagation

SkeleTR: Towards Skeleton-based Action Recognition in the Wild

Exploiting weakly supervised visual patterns to learn from partial annotations

Id-Free Person Similarity Learning

PSS: Progressive Sample Selection for Open-World Visual Representation Learning

Learning for Transductive Threshold Calibration in Open-World Recognition