Multi-modal art generation and capitalization method based on emotion recognition

Abstract

The invention discloses a multi-modal art generation and emotional asset system based on emotional coordinates, which recognizes main emotions and maps coordinates through data such as images, keywords, music and expressions input by a user, matches colors, styles and content templates in combination with a strategy library, generates multi-modal works such as emotional graph cards, music, ornaments and life suggestions, and provides a multi-modal art generation and emotional asset system based on the emotional coordinates. NFT and other digital emotional assets can be packaged, user collection and transaction are supported, and meanwhile the system records user feedback optimization recommendation.

Description

1. Technical Field

[0001] The present invention relates to the fields of artificial intelligence, human-computer interaction, art generation and digital assets, and in particular to a multimodal art generation and assetization method and system based on emotion recognition, which can be applied to scenarios such as personalized art creation, psychological healing, digital collections and lifestyle recommendations. 2. Background Technology

[0002] Emotion recognition technology is currently being used in areas such as health monitoring and personalized recommendations, but most systems only provide label-based output and lack structured, multimodal visual, audible, and interactive output. Furthermore, while art generation and NFT digital asset technologies are rapidly developing, user engagement with subjective emotions is low, making it difficult to establish a personalized connection between internal emotions and external content. Therefore, there is an urgent need for a multimodal generation and asset system based on emotional coordinates that can transform user emotions into images, music, accessories, and even lifestyle recommendations through structured mapping, thereby enhancing freedom of expression and the ability to transform emotional value. 3. Summary of the Invention

[0003] The present invention provides a system and method for converting a user's multi-source emotional input into multimodal artistic expression and digital assets. The system includes a user input module, an emotion recognition and coordinate mapping module, a strategy matching engine, a multimodal generation module, a visualization module, an emotion asset packaging module, and a user feedback mechanism. The system supports multi-source emotional input such as images, keywords, music, behavior descriptions, expressions, and living environments. After mapping to emotional coordinates, it calls style, color, music, makeup, accessories, and life matching templates to generate personalized picture cards, music clips, art paintings, outfit / makeup / space suggestions, and other content, which can be packaged as NFT assets. The final user behavior data is fed back to the strategy system to achieve a closed loop of content optimization. IV. Description of the Figures

[0004] The accompanying figure is a flow chart of the overall structure of the system described in the present invention, including: ① user input stage, ② emotion recognition and coordinate mapping, ③ generation strategy matching, ④ multimodal generation, ⑤ product visualization, ⑥ emotional asset generation (optional), and ⑦ user feedback closed loop. V. Specific Implementation Methods

[0005] Users can upload a selfie, select emotional keywords, and describe their behavior. The system will identify their primary emotion, such as "calm + joyful," and automatically match it to coordinates, including impressionist painting styles, soft tones, a medium-tempo piano music template, a sketch of a water droplet-shaped accessory, and light-colored, transparent outfit suggestions. These are then displayed as "emotional artworks" within the interface. Users can choose whether to generate these artworks as NFTs for collection, display, or social sharing. The system will record user clicks, collections, and preferred creations, continuously optimizing the generation strategy and emotion matching model.

Claims

1. A multimodal art generation and assetization method based on emotion recognition, characterized by: The steps include: (1) receiving emotion-related information input by the user, including images, keywords, music clips, expression data, behavior descriptions, images of living environments, and the user's favorite colors or paintings. The information can be uploaded by the user or selected from the system database; (2) identifying the main emotion of the above information and mapping it to a position in a preset two-dimensional or polar coordinate system; (3) Matching content strategies including color style, music style template, art style, accessory structure, makeup suggestions, clothing suggestions, space decoration suggestions, and lifestyle accessories matching suggestions in the strategy library according to the coordinate position; (4) outputting at least one multimodal artistic expression such as an emotion card, a music clip, an art painting, a jewelry pattern, a DIY jewelry bag, or a lifestyle suggestion through a multimodal generation module according to the matching strategy; (5) Visually display the above generated content to form a unified emotional work interface; (6) Optionally package the emotional works into digital emotional assets to generate NFTs or other tradable digital forms; (7) Record user behavior preferences, collection behavior and other data, and use them to optimize generation algorithms and recommendation strategies.

2. The method according to claim 1, wherein the emotion coordinate system is a two-dimensional plane or a polar coordinate diagram, and the coordinate axes represent pleasure, arousal or other combinations of emotion dimensions. The method according to claim 1 , wherein the music style template supports a user to specify a composer or a music period.

4. The method according to claim 1, wherein the generated jewelry design pattern can be used for 3D modeling or DIY material package output.

5. The method according to claim 1, wherein the life suggestions include makeup suggestions, clothing suggestions, space decoration styles or life accessories combination plans.

6. The method according to claim 1, wherein the digital emotional assets include NFT images, music, jewelry samples, life advice cards, etc.

Images