The trend of research on the 3D animation of static art works is increasing, and the 3D animation of traditional Chinese art works has gradually become a hot topic in this type of research. Through image segmentation, image repair and animation synthesis technology, this study has carried out artistic 3D animation generation of landscape painting in the Northern Song Dynasty. Firstly, three image segmentation methods, BMD&OTSU, SAM and Fast SAM, are used to segment multilayered painting, the foreground and background of landscape; Secondly, the segmentation results are used as masks to fill the background and extract the foreground of the original image; Finally, render and animate the filled landscape painting. This study has found a technology roadmap that can realize the 3D animation of static painting, and has provided a way of thinking for the remodeling of traditional Chinese art works.

Fashion trend forecasting is crucial for proactive supply chain management, sustainable production, and personalized marketing. However, accurately predicting fine-grained fashion trends across diverse user groups remains challenging, as conventional numerical models rely solely on historical popularity signals and cannot explicitly exploit the high-level semantic context underlying trend evolution. To address this limitation, we propose a Dual-Stream Driven Fashion Trend Forecasting (DDFTF) framework that integrates numerical forecasting with textual semantic reasoning. The numerical stream employs a multi-scale patch Transformer with metadata-aware feature fusion to capture temporal dynamics, while the semantic stream converts historical trends into structured textual descriptions, extracting interpretable semantic representations of global trends and turning points. A residual fusion module combines both streams by treating semantic signals as complementary corrections to numerical predictions. Extensive experiments on the FIT and GeoStyle benchmarks demonstrate that DDFTF consistently outperforms state-of-the-art methods, achieving up to a 14.6% relative MAE reduction in long-term forecasting. Ablation and qualitative analyses further show that the performance gains arise from meaningful semantic information rather than increased model capacity, while also improving the interpretability of fashion trend prediction. Our code is publicly available.