Research on Style Innovation and Precise Audience Targeting Path of Visual Communication of Animation IP Empowered by Deep Learning

Visual communication of animation IP is an important link between content creation and audience reception. Under the development of intelligent production technologies, animation design is shifting from traditional manual production toward data-driven intelligent creation. This paper analyzes visual style innovation and precise audience targeting in animation IP communication by integrating deep learning technologies, including generative adversarial networks, convolutional neural networks, and reinforcement learning. An intelligent creation and communication model is developed, including a multidimensional animation style generation system, an audience profiling method, and a propagation path optimization mechanism. The animation style generation model extracts visual features, analyzes style attributes, and generates new content based on controllable parameters. Audience profiles are constructed through multi-source user behavior data, while dissemination paths are optimized using reinforcement learning. Experimental results show that deep learning improves the diversity, consistency, and visual quality of animation IP expression, enhances communication efficiency, and supports more accurate audience reach through personalized content recommendation.