Verified80% confidenceFactExact time
IP-Adapter (Image Prompt Adapter) was proposed by Tencent Research in 2023; it preserves character identity features by encoding the visual features of a reference image and injecting them into the cross-attention layers of a diffusion model.
4
Sources
80%
Confidence
Long-term
Relevance
7/12/2026
First Seen
Sources
AI-Generated D&D Fantasy Character Portraits: A New Creative Tool for Tabletop Gamers
redditr/midjourney7/10/2026
Related Claims
VerifiedIP-Adapter (Image Prompt Adapter) was introduced by Tencent ARC Lab and can transfer the style and character features of a reference image to the generated output.84% similarVerifiedIP-Adapter, proposed by Tencent Research Institute, converts character appearance into image prompt vectors through an independent image encoder, enabling appearance feature transfer without modifying model weights.81% similarUnverifiedIP-Adapter is a technique that injects the semantic features of a reference image into the generation process through a decoupled cross-attention mechanism, enabling simultaneous guidance from image features and text prompts.77% similarUnverifiedIP-Adapter allows users to directly input a reference image as a visual prompt to quickly reproduce a character's appearance without training a LoRA.73% similarUnverified实现跨镜头角色一致性的技术手段包括参考图像注入（如IP-Adapter）、LoRA微调、以及对话式AI的上下文记忆能力72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/490909API
curl https://kongchang.com/api/v1/knowledge/claims/490909MCP
get_claim(id=490909)