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

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.

Authors

• Zhicheng Li
• Kunyang Sun China Unviersity of Mining and Technology
• Rui Yao
• Hancheng Zhu China Unviersity of Mining and Technology - Xuzhou Mine Digitization Engineering Research Center of the Ministry of Education - Xuzhou
• Fuyuan Hu School of Electronic and Information Engineering, Suzhou University of Science and Technology, China Suzhou Key Laboratory of Intelligent Low Carbon Technology Application, China Jiangsu Industrial Intelligent Low Carbon Technology Engineering Center, China
• Jiaqi Zhao China Unviersity of Mining and Technology - Xuzhou Mine Digitization Engineering Research Center of the Ministry of Education - Xuzhou
• Zhiwen Shao The Hong Kong University of Science and Technology China University of Mining Technology
• Yong Zhou ShanghaiTech University

Keywords

No keywords are indexed for this paper.