Xinyu Li

papers

911

total citations

papers (18)

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

CVPR 2022arXiv

citations

Revisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokens

CVPR 2023arXiv

citations

Directional Temporal Modeling for Action Recognition

ECCV 2020arXiv

citations

Motion-Guided Masking for Spatiotemporal Representation Learning

ICCV 2023arXiv

citations

Multi-Label Activity Recognition Using Activity-Specific Features and Activity Correlations

CVPR 2021arXiv

citations

Stochastic Backpropagation: A Memory Efficient Strategy for Training Video Models

CVPR 2022arXiv

citations

Selective Feature Compression for Efficient Activity Recognition Inference

ICCV 2021arXiv

citations

Text-Guided Video Masked Autoencoder

ECCV 2024arXiv

citations

MEGA: Multimodal Alignment Aggregation and Distillation For Cinematic Video Segmentation

ICCV 2023arXiv

citations

DZAD: Diffusion-based Zero-shot Anomaly Detection

AAAI 2025

citations

PAC-Bayes Bounds for Multivariate Linear Regression and Linear Autoencoders

NEURIPS 2025arXiv

citations

Id-Free Person Similarity Learning

CVPR 2022

citations

Debiased Causal Tree: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding

NEURIPS 2022

citations

SemiVisBooster: Boosting Semi-Supervised Learning for Fine-Grained Classification through Pseudo-Label Semantic Guidance

ICCV 2025

citations

Xinyu Li

papers (18)

VidTr: Video Transformer Without Convolutions

Long Short-Term Transformer for Online Action Detection

SiamMOT: Siamese Multi-Object Tracking

TubeR: Tubelet Transformer for Video Action Detection

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

Revisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokens

Directional Temporal Modeling for Action Recognition

Motion-Guided Masking for Spatiotemporal Representation Learning

Multi-Label Activity Recognition Using Activity-Specific Features and Activity Correlations

Stochastic Backpropagation: A Memory Efficient Strategy for Training Video Models

Selective Feature Compression for Efficient Activity Recognition Inference

Text-Guided Video Masked Autoencoder

MEGA: Multimodal Alignment Aggregation and Distillation For Cinematic Video Segmentation

DZAD: Diffusion-based Zero-shot Anomaly Detection

PAC-Bayes Bounds for Multivariate Linear Regression and Linear Autoencoders

Id-Free Person Similarity Learning

Debiased Causal Tree: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding

SemiVisBooster: Boosting Semi-Supervised Learning for Fine-Grained Classification through Pseudo-Label Semantic Guidance

papers (18)

VidTr: Video Transformer Without Convolutions

Long Short-Term Transformer for Online Action Detection

SiamMOT: Siamese Multi-Object Tracking

TubeR: Tubelet Transformer for Video Action Detection

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

Revisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokens

Directional Temporal Modeling for Action Recognition

Motion-Guided Masking for Spatiotemporal Representation Learning

Multi-Label Activity Recognition Using Activity-Specific Features and Activity Correlations

Stochastic Backpropagation: A Memory Efficient Strategy for Training Video Models

Selective Feature Compression for Efficient Activity Recognition Inference

Text-Guided Video Masked Autoencoder

MEGA: Multimodal Alignment Aggregation and Distillation For Cinematic Video Segmentation

DZAD: Diffusion-based Zero-shot Anomaly Detection

PAC-Bayes Bounds for Multivariate Linear Regression and Linear Autoencoders

Id-Free Person Similarity Learning

Debiased Causal Tree: Heterogeneous Treatment Effects Estimation with Unmeasured Confounding

SemiVisBooster: Boosting Semi-Supervised Learning for Fine-Grained Classification through Pseudo-Label Semantic Guidance