Yunhai Tong

Latest

• Explore In-Context Segmentation via Latent Diffusion Models
• Towards robust referring image segmentation
• Sfnet: Faster and accurate semantic segmentation via semantic flow
• Towards open vocabulary learning: A survey
• RAP-SAM: Towards Real-Time All-Purpose Segment Anything
• Towards language-driven video inpainting via multimodal large language models
• Mitigating Semantic Confusion from Hostile Neighborhood for Graph Active Learning
• Dst-det: Simple dynamic self-training for open-vocabulary object detection
• Multiple Connectivity Views for Session-based Recommendation
• Betrayed by captions: Joint caption grounding and generation for open vocabulary instance segmentation
• Panopticpartformer++: A unified and decoupled view for panoptic part segmentation
• Convolution-enhanced evolving attention networks
• Label-efficient interactive time-series anomaly detection
• TransVOD: End-to-end Video Object Detection with Spatial-Temporal Transformers
• Panoptic-partformer: Learning a unified model for panoptic part segmentation
• Polyphonicformer: Unified query learning for depth-aware video panoptic segmentation
• Fashionformer: A simple, effective and unified baseline for human fashion segmentation and recognition
• Query Learning of Both Thing and Stuff for Panoptic Segmentation
• Improving Video Instance Segmentation via Temporal Pyramid Routing
• Enhancing self-attention with knowledge-assisted attention maps
• Heat-RL: Online Model Selection for Streaming Time-Series Anomaly Detection
• Graph pointer neural networks
• Ts2vec: Towards universal representation of time series
• Video k-net: A simple, strong, and unified baseline for video segmentation
• BoundarySqueeze: Image Segmentation as Boundary Squeezing
• Improving BERT with Self-Supervised Attention
• End-to-end video object detection with spatial-temporal transformers
• Dynamic Dual Sampling Module For Fine-Grained Semantic Segmentation
• Fast and accurate scene parsing via bi-direction alignment networks
• Global aggregation then local distribution for scene parsing
• Competence-based Curriculum Learning for Multilingual Machine Translation
• Towards efficient scene understanding via squeeze reasoning
• Evolving attention with residual convolutions
• Customizing Graph Neural Networks using Path Reweighting
• PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation
• Enhanced boundary learning for glass-like object segmentation
• Fast and Accurate Scene Parsing via Bi-Direction Alignment Networks
• Syntax-BERT: Improving Pre-trained Transformers with Syntax Trees
• Multivariate time-series anomaly detection via graph attention network
• AutoADR: Automatic model design for ad relevance
• Spectral temporal graph neural network for multivariate time-series forecasting
• Boundary content graph neural network for temporal action proposal generation
• Improving semantic segmentation via decoupled body and edge supervision
• Semantic flow for fast and accurate scene parsing
• Ladabert: Lightweight adaptation of bert through hybrid model compression
• Gated fully fusion for semantic segmentation
• Spherical criteria for fast and accurate 360 object detection
• Textnas: A neural architecture search space tailored for text representation
• Customized graph embedding: tailoring embedding vectors to different applications
• Global aggregation then local distribution in fully convolutional networks
• Dual graph convolutional network for semantic segmentation
• Flow2seg: Motion-aided semantic segmentation
• Reprojection R-CNN: A Fast and Accurate Object Detector for 360 deg Images
• A cooperative multi-agent reinforcement learning framework for resource balancing in complex logistics network
• Fast parallel path concatenation for graph extraction
• Gvos: a general system for near-duplicate video-related applications on storm
• A Structural Based Community Similarity Algorithm and Its Application in Scientific Event Detection
• Netrating: Credit risk evaluation for loan guarantee chain in china
• StroMAX: Partitioning-based scheduler for real-time stream processing system
• Ranking scientific articles by exploiting citations, authors, journals, and time information