发表机构
Peking University（北京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对产品中心广告视频生成缺乏大规模数据和评估框架的问题，提出AdSpark数据集与基准，含30万三元组和六维评估，验证了数据集有效性并揭示关键挑战。
AI 中文摘要
产品中心广告视频生成旨在创建能够保留细粒度产品身份，同时通过连贯的多镜头叙事呈现卖点的促销视频。然而，由于缺乏大规模广告专用数据集和全面的评估框架，这一新兴任务仍未得到充分探索。为弥补这一空白，我们基于一家大型电商平台的数据，推出了 AdSpark，一个用于产品中心广告视频生成的大规模数据集和基准。AdSpark-300K 包含约 30 万个参考图像-提示-视频三元组，涵盖真实世界子集和合成子集。每个样本提供结构化的广告标注，包括产品身份标注、卖点描述、创意方案和对齐的音频脚本，使模型能够学习产品保留和面向广告的视觉叙事。我们进一步提出了 AdSpark-Bench，一个诊断基准，从六个维度评估生成的广告，包括视觉质量、产品保真度、指令遵循、时间连贯性、音频对齐和广告效果。基于 AdSpark-Bench，我们评估了代表性模型，揭示了产品保留、多镜头叙事和卖点可视化方面的关键挑战。使用 AdSpark-300K 微调模型的实验进一步验证了我们数据集的有效性。AdSpark 为未来研究提供了统一的数据集和基准，我们将在论文被接收后发布该数据集。
英文摘要
Product-centric advertisement video generation aims to create promotional videos that preserve fine-grained product identity while presenting selling points through coherent multi-shot narratives. However, this emerging task remains underexplored due to the lack of large-scale advertisement-specific datasets and comprehensive evaluation frameworks. To address this gap, we introduce \textbf{AdSpark}, a large-scale dataset and benchmark for product-centric advertisement video generation, based on data from a major e-commerce platform. \textit{AdSpark-300K} contains approximately 300K reference image--prompt--video triplets, comprising a real-world subset and a synthetic subset. Each sample provides structured advertisement annotations, including product identity annotations, selling-point descriptions, creative plans, and aligned audio scripts, enabling models to learn product preservation and advertisement-oriented visual storytelling. We further propose \textit{AdSpark-Bench}, a diagnostic benchmark that evaluates generated advertisements across six dimensions, including visual quality, product fidelity, instruction adherence, temporal coherence, audio alignment, and advertisement effectiveness. Based on AdSpark-Bench, we evaluate representative models, revealing key challenges in product preservation, multi-shot storytelling, and selling-point visualization. Experiments with AdSpark-300K-finetuned models further validate the effectiveness of our dataset. AdSpark provides a unified dataset and benchmark for future research, and we will release the dataset upon acceptance.
