ACM 国际多媒体大会（英文名称：ACM Multimedia，简称：ACM MM）是多媒体领域的顶级国际会议，每年举办一次。
VIP内容
近年来，“AI 变脸”特效风靡全球，近期爆红的 “蚂蚁呀嘿” 再次掀起体验和讨论的热潮，这种源自人工智能人脸生成的新技术，能够利用深度学习技术识别并替换图片或视频中的原始人像，不仅制作过程逐渐简单化，而且逼真度惊人，几乎能达到以假乱真的效果。
Deepfake作为一项技术工具，有着广泛的应用空间。语音合成能让计算机用人类的声音说出上百种语言，视频合成能让《速度与激情》里的 Paul Walker 复生，但若被滥用，也将带来巨大的风险，给身份识别和社会信任带来挑战，虚假视觉信息的应用与传播还会给人们造成隐私安全等多方面的困扰。
随着DeepFake等人脸生成技术的发展，伪造的人脸越来越逼真，有些甚至达到人眼也难以区分真假的地步。对此，腾讯优图实验室提出了一种全新的鉴伪方法，该方法同时从时间和空间不一致性入手, 对视频人脸伪造进行鉴别，这将有助于鉴别社交网络上传播的虚假视频，守护人脸安全。目前，该方法在四个学术基准数据集上均取得领先结果，相关论文已被多媒体领域会议ACM MM 2021收录。
论文下载地址：
https://www.zhuanzhi.ai/paper/b796509ba07ab7d0901ae85ba76128de
成为VIP会员查看完整内容
Spatiotemporal Inconsistency Learning for DeepFake Video Detection
热门内容
最新内容
To unfold the tremendous amount of multimedia data uploaded daily to social media platforms, effective topic modeling techniques are needed. Existing work tends to apply topic models on written text datasets. In this paper, we propose a topic extractor on video transcripts. Exploiting neural word embeddings through graph-based clustering, we aim to improve usability and semantic coherence. Unlike most topic models, this approach works without knowing the true number of topics, which is important when no such assumption can or should be made. Experimental results on the real-life multimodal dataset MuSe-CaR demonstrates that our approach GraphTMT extracts coherent and meaningful topics and outperforms baseline methods. Furthermore, we successfully demonstrate the applicability of our approach on the popular Citysearch corpus.
最新论文
To unfold the tremendous amount of multimedia data uploaded daily to social media platforms, effective topic modeling techniques are needed. Existing work tends to apply topic models on written text datasets. In this paper, we propose a topic extractor on video transcripts. Exploiting neural word embeddings through graph-based clustering, we aim to improve usability and semantic coherence. Unlike most topic models, this approach works without knowing the true number of topics, which is important when no such assumption can or should be made. Experimental results on the real-life multimodal dataset MuSe-CaR demonstrates that our approach GraphTMT extracts coherent and meaningful topics and outperforms baseline methods. Furthermore, we successfully demonstrate the applicability of our approach on the popular Citysearch corpus.
