Detection of AI-Generated Product Main Images in Cross-Border E-Commerce Based on Multi-Feature Fusion
Abstract
Abstract With the rapid development of generative artificial intelligence-generated content (AIGC), diffusion models such as Midjourney, Stable Diffusion, and DALL-E 3 have increasingly been used to produce product main images for cross-border e-commerce. These tools can reduce the cost of visual content production for sellers; however, AI-generated product images may still contain artifacts in micro-texture continuity, object boundaries, illumination consistency, refraction, and high-order statistical distributions. Such artifacts may increase the risk of misleading visual presentation and create new challenges for platform image auditing. Existing general-purpose AIGC detection methods, digital watermarking techniques, and Deepfake identification models may not transfer well to product main images in cross-border e-commerce. A major reason is that such images often contain dense promotional text, multilingual layout watermarks, multi-scale price labels, and decorative backgrounds. These elements introduce strong edge and texture interference, making it difficult for conventional global feature extractors to distinguish artificial layout features from generation artifacts. To address this domain-specific problem, this study constructs CBEC-AIGC-Bench, a benchmark dataset covering four cross-border e-commerce categories: digital electronics, apparel and footwear, marine fresh food, and outdoor fishing equipment. A multi-feature fusion network, MFF-EComDet, is then proposed…
Read the paper
Related studies
- Learning from First-Generation Qualitative Approaches in the IS Discipline: An Evolutionary View and Some Implications for Authors and Evaluators (PART 1/2)
- Commentary on the clinical and preclinical dosage limits of interstitially administered magnetic fluids for therapeutic hyperthermia based on current practice and efficacy models
- Prosthesis: New Technological Opportunities and Innovative Biomedical Devices
- The history and future of neural modeling for cochlear implants
- Ryanodine receptors: waking up from refractoriness
- Novel Transgenic Rabbit Model Sheds Light on the Puzzling Role of Matrix Metalloproteinase-12 in Atherosclerosis
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.