DTTNet: Improving Video Shadow Detection via Dark-Aware Guidance and Tokenized Temporal Modeling

TL;DR

DTTNet introduces a novel approach for video shadow detection that combines linguistic priors, adaptive reweighting, and tokenized temporal modeling to improve accuracy and efficiency in complex scenes.

Contribution

The paper proposes DTTNet, which integrates dark-aware guidance and tokenized temporal modeling for enhanced video shadow detection, achieving state-of-the-art results.

Findings

State-of-the-art accuracy on benchmark datasets

Real-time inference efficiency

Effective differentiation of shadows from dark objects

Abstract

Video shadow detection confronts two entwined difficulties: distinguishing shadows from complex backgrounds and modeling dynamic shadow deformations under varying illumination. To address shadow-background ambiguity, we leverage linguistic priors through the proposed Vision-language Match Module (VMM) and a Dark-aware Semantic Block (DSB), extracting text-guided features to explicitly differentiate shadows from dark objects. Furthermore, we introduce adaptive mask reweighting to downweight penumbra regions during training and apply edge masks at the final decoder stage for better supervision. For temporal modeling of variable shadow shapes, we propose a Tokenized Temporal Block (TTB) that decouples spatiotemporal learning. TTB summarizes cross-frame shadow semantics into learnable temporal tokens, enabling efficient sequence encoding with minimal computation overhead. Comprehensive Experiments on multiple benchmark datasets demonstrate state-of-the-art accuracy and real-time inference efficiency. Codes are available at https://github.com/city-cheng/DTTNet.