Tianyu (Steve) Wang

An MLLM-based agent framework for instruction-driven image retouching with a generalist reward model.

A diffusion Transformer framework for multi-subject video customization with multimodal control conditions, featuring a data construction pipeline for training without labels.

A comprehensive survey and benchmark covering the past decade of shadow detection, removal, and generation research in deep learning.

A semi-supervised framework for video instance shadow detection using contrastive learning and cycle consistency.

A unified framework for image restoration that learns both weather-general and weather-specific features to handle multiple adverse conditions.

A network for real-time 3D hand mesh reconstruction that handles hand occlusion and orientation challenges.

A method for learning shadow detection from noisy labels using iterative label tuning with shadow-aware mechanisms.

Journal extension of CVPR 2021 work with improved single-stage detector for instance shadow detection and shadow-object association.

A framework to boost 3D detection by learning to densify point clouds in latent space, improving detection of small and distant objects.

A single-stage approach for detecting instance-level shadows using bidirectional relation learning between objects and their shadows.

A comprehensive benchmark dataset for shadow detection in complex real-world scenarios.

First work to tackle instance-level shadow detection, associating each shadow with its corresponding object instance.

Real-time rain removal using depth-guided non-local features for robust image restoration.

Spatial attention mechanism for salient object detection using context attenuation.

Spatial attention network for single-image deraining with a new high-quality real-world rain dataset.

Real-time 3D hand mesh reconstruction with improved alignment accuracy.