Self-Reasoning Agentic Framework for Narrative Product Grid-Collage Generation

2026-04-18

AI Summary

Existing image generation methods often lack structured narrative planning and cross-frame coordination, resulting in product collage sequences that exhibit weak storytelling and visual incoherence. This work proposes the first self-reflection agent framework tailored for narrative-driven product collage generation. By explicitly modeling narrative logic through product-centric story structure construction, constraint-aware prompt compilation, and a unified multi-frame image generation model, the approach jointly synthesizes coherent visual sequences. Furthermore, it introduces a dual-dimensional gating mechanism evaluating both content fidelity and photographic quality to enable failure attribution–guided iterative refinement. Compared to direct prompting baselines, the proposed method significantly enhances the aesthetic quality, narrative richness, and visual consistency of generated collages.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Narrative-driven product photography has become a prevalent paradigm in modern marketing, as coherent visual storytelling helps convey product value and establishes emotional engagement with consumers. However, existing image generation methods do not support structured narrative planning or cross-panel coordination, often resulting in weak storytelling and visual incoherence. In practice, narrative product photography is commonly presented as multi-grid collages, where multiple views or scenes jointly communicate a product narrative. To ensure visual consistency across grids and aesthetic harmony of the overall composition, we generate the collage as a single unified image rather than composing independently synthesized panels. We propose a self-reasoning agentic framework for narrative product grid collage generation. Given a product packshot and its name, the system first constructs a Product Narrative Framework that explicitly represents the product's identity, usage context, and situational environment, and translates it into complementary grids governed by a shared visual style. Constraint-aware prompts are then compiled and fed to a generation model that synthesizes the collage jointly. The generated output is evaluated on both content validity and photography quality, with explicit gates determining whether to proceed or refine. When evaluation fails, the system performs failure attribution and applies targeted refinement, enabling progressive improvement through iterative self-reflection. Experiments demonstrate that our framework consistently improves aesthetic quality, narrative richness, and visual coherence, compared to direct prompting baselines.

Problem

• narrative product photography
• grid collage generation
• visual coherence
• structured narrative planning
• cross-panel coordination

Innovation

• self-reasoning
• agentic framework
• narrative product photography
• grid collage generation
• constraint-aware prompting