How AI Revolutionized the “One Orange” IP Design Workflow

Why Switch to an AI‑Powered Production Flow

Business drivers: large‑scale events and long‑term incentive campaigns rely on IP as a visual symbol to boost product recognition, create memorable touchpoints, and generate reusable assets that cut costs and increase efficiency.

Team drivers: the classic IP pipeline required many steps, heavy 3D staffing, and long cycles, making rapid iteration impossible. Starting with the 2025 Gaokao project, the Search design team set a year‑long goal to rebuild IP production with AI, aiming to meet material volume and quality while breaking scene constraints and reducing resource dependence.

Technical drivers: AI tools are not the end of design but the beginning of a cognitive shift. When AI becomes a standard part of the workflow, the challenge is to let technology serve core creative needs without stifling originality or emotional warmth.

AI‑Enabled Dual Breakthroughs: Innovation and Efficiency

Innovation Breakthrough

Process innovation: moving from a linear "2D illustration → three‑view → 2D‑to‑3D → static‑to‑dynamic → multi‑platform adaptation" pipeline to a modular, parallel‑task workflow powered by AI. Designers break tasks into modules, allowing simultaneous production and guaranteeing style consistency across all stages.

Technical innovation: replacing manual drawing with multimodal generation. Text prompts are turned into visual concepts, then automatically fed into 3D modeling, animation, and derivative material pipelines.

Efficiency Breakthrough

From "limited trial‑and‑error" to "near‑infinite possibilities": AI can compress early creative exploration by over 90%, generating thousands of style variations and animation demos in minutes, enabling rapid high‑potential concept selection.

From "repetitive labor" to "precise high‑efficiency": AI‑driven one‑click 2D‑to‑3D, static‑to‑dynamic, and text‑to‑video conversion eliminates costly manual steps. By training a dedicated LoRA model and using ControlNet to lock core proportions, colors, and textures, the system can batch‑produce compliant assets, boosting productivity by more than tenfold.

Full‑Process AI Workflow Highlights

Highlight 1 – LoRA‑Based AI Generation : AI creates raw frames, which are post‑processed into high‑quality IP samples (e.g., 30 core character images). These samples train a custom LoRA model that locks the IP’s proportions, facial features, colors, and texture, enabling rapid, consistent generation for concept expansion and multi‑media adaptation.

Highlight 2 – LoRA + ControlNet Precise Multi‑Dimensional Control : Combining LoRA with ControlNet achieves ≤2% error when mapping core IP images to control maps (skeleton, depth, pose, expression). First‑frame prompts and standardized prompts stabilize output, delivering fluid and nuanced animation.

Highlight 3 – AI‑Powered Multi‑Style Transfer : A style‑transfer model takes the core IP image and a target style reference, instantly converting the character into various visual styles (e.g., 3D → sketch). The process preserves core features, ensuring consistent branding while supporting rapid material iteration.

IP Application Scenarios

The “One Orange” IP appears across 2025 core activities—Gaokao promotion, summer earning walks, team‑earning events, back‑to‑school campaigns—demonstrating full‑chain AI‑enabled IP penetration that lifts overall efficiency by over 50% while maintaining visual quality.

Design Reflection: Co‑Creating with AI to Maximize Human Value

AI‑design synergy is a two‑way partnership: technology empowers creativity while human insight anchors purpose. AI liberates designers from repetitive execution, allowing focus on user research, concept ideation, and value creation.

Despite hype that AI will replace designers, the article argues that AI can only automate the "technique" layer—standardized, repeatable tasks—while the "philosophy" layer—insight, aesthetic judgment, and cultural relevance—remains uniquely human.

Designers must therefore anchor their irreplaceability by embracing AI as a tool, keeping humanity at the core, and continuously advancing creative thinking in the AI era.