King Abdullah University of Science and Technology (KAUST)(阿卜杜拉国王科技大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Beijing Zhongguancun Academy(北京中关村学院);Zhongguancun Institute of Artificial Intelligence(中关村人工智能研究院)
交互式世界模型必须捕捉多个参与者在观察共享环境的不同部分时如何共同改变该环境的过程,竞技游戏将这一挑战具象化:玩家既合作又竞争,而战争迷雾、团队视野和相机位置决定了每个玩家能看到的内容。对这些交互进行建模需要能将个体观察与动作关联到共享状态和事件的数据。我们推出电竞世界模型数据集(Esports World Model Dataset,EWM),该数据集包含《星际争霸II》《Dota 2》和《反恐精英2》的人类游戏玩法,共3053场游戏及地图,约4880小时的玩家视角时长,其中包含正在进行的《反恐精英2》扩展内容的预估时长。它包含4628个已完成的《星际争霸II》视角、303场《Dota 2》比赛的全部10个视角,以及《反恐精英2》每回合最多10个视角。在这些视角中,回放时钟对齐将渲染视频与同一场比赛的原生游戏动作、结构化状态及事件关联起来。三款游戏呈现出互补的观察条件:私有战争迷雾、共享团队视野和第一人称视角。在《星际争霸II》中,玩家特定的实体记录进一步区分了实时观察与记忆快照,并关联了对立视角间共同观察到的实体。结合时间对齐,这些记录使研究人员能够区分合法信息差异与共享状态的分歧。通过关联人类决策、共享动态和玩家特定信息,EWM支持朝着联合控制的多人世界模型以及由人类游戏学习得到的队友和对手组成的交互式环境开展研究。
英文摘要:
Interactive world models must capture how multiple participants jointly change a shared environment while observing different parts of it. Competitive games make this challenge concrete: players cooperate and compete, while fog of war, team vision and camera placement determine what each can see. Modeling these interactions requires data that connect individual observations and actions to shared state and events. We introduce the Esports World Model Dataset (EWM), a collection of human gameplay spanning StarCraft II, Dota 2 and Counter-Strike 2, with 3,053 games and maps and approximately 4,880 player-view hours, including estimated hours for the ongoing CS2 extension. It includes 4,628 completed StarCraft II views, all ten perspectives for 303 Dota 2 matches, and up to ten views per round in CS2. Across these views, replay-clock alignment connects rendered video with game-native actions, structured state and events from the same match. The three games expose complementary observation conditions: private fog of war, shared team vision and first-person viewpoints. In StarCraft II, player-specific entity records further distinguish live observations from remembered snapshots and link co-observed entities across opposing views. Together with temporal alignment, these records allow researchers to distinguish legitimate information differences from disagreement about shared state. By connecting human decisions, shared dynamics and player-specific information, EWM supports research toward jointly controlled multiplayer world models and interactive environments populated by teammates and opponents learned from human play.
DrivingDojo Dataset: Advancing Interactive and Knowledge-Enriched Driving World Model
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Meituan Inc.(美团公司);Centre for Artificial Intelligence and Robotics, HKISI, CAS(中国科学院香港智能研究院人工智能与机器人中心);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院)
Shanghai AI Laboratory(上海人工智能实验室);Beijing Institute of Technology(北京理工大学);Shanghai Innovation Institute(上海创新研究院);Shenzhen MSU-BIT University(深圳MSU-BIT大学);The University of Tokyo(东京大学)
Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data
Hong Kong University of Science and Technology(香港科技大学);Singapore University of Technology and Design(新加坡科技设计大学);The University of Tokyo(东京大学);Deakin University(迪肯大学)
Department of Computing, The Hong Kong Polytechnic University(香港理工大学计算机系);School of Science and Engineering, The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳)理工学院)
SmokeBench: A Real-World Dataset for Surveillance Image Desmoking in Early-Stage Fire Scenes
Beijing Jiaotong University(北京交通大学);Dalian Polytechnic University(大连理工大学);Nanjing University of Aeronautics and Astronautics(南京航空航天大学);Dalian Martime University(大连海洋大学);Nanjing University of Science and Technology(南京理工大学)
MFFI: Multi-Dimensional Face Forgery Image Dataset for Real-World Scenarios
Anhui Province Key Laboratory of Digital Security(安徽省级数字安全重点实验室);Hefei University of Technology(合肥工业大学);Cardiff University(卡迪夫大学);IHPC, Agency for Science, Technology and Research(科技研究局信息处理中心);Beijing Academy of Artificial Intelligence(北京人工智能研究院)
Towards Real-World Deepfake Detection: A Diverse In-the-wild Dataset of Forgery Faces
School of Cyber Science and Engineering, Huazhong University of Science and Technology(华中科技大学信息科学与工程学院);School of Software Engineering, Huazhong University of Science and Technology(华中科技大学软件工程学院);School of Information and Communication Technology, Griffith University(格里菲斯大学信息与通信技术学院);School of Computer Science and Technology, Huazhong University of Science and Technology(华中科技大学计算机科学与技术学院)
Super-resolving Real-world Image Illumination Enhancement: A New Dataset and A Conditional Diffusion Model
The Chinese University of Hong Kong(香港中文大学);City University of Hong Kong(香港城市大学);Nanjing University of Science and Technology(南京理工大学);Lingnan University(岭南大学)
V2V4Real: A Real-world Large-scale Dataset for Vehicle-to-Vehicle Cooperative Perception
University of California Los Angeles(加州大学洛杉矶分校);Cleveland State University(克利夫兰州立大学);University of Texas at Austin(德克萨斯大学奥斯汀分校);Northwestern University(西北大学);Technical University of Munich(慕尼黑工业大学);Fraunhofer Institute(弗劳恩霍夫研究所)
自动驾驶车辆的现代感知系统已知对遮挡敏感并且缺乏长感知范围的能力。它一直是阻碍Level 5自治的关键瓶颈之一。最近的研究表明,车对车(V2 V)协同感知系统具有巨大的潜力,可以彻底改变自动驾驶行业。然而,缺乏真实世界的数据集阻碍了这一领域的进展。为了促进合作感知的发展,我们提出了V2 V4 Real,这是第一个用于V2 V感知的大规模真实世界多模态数据集。数据由两辆配备多模态传感器的车辆收集,它们在不同的场景中一起行驶。我们的V2 V4 Real数据集覆盖了410公里的驾驶区域,包括20 K LiDAR帧,40 K RGB帧,240 K注释的5类3D边界框,以及覆盖所有驾驶路线的HDMap。V2 V4 Real介绍了三种感知任务,包括协作3D目标检测、协作3D对象跟踪和Sim 2 Real用于协作感知的领域自适应。我们提供了全面的基准最近的合作感知算法的三个任务。V2 V4 Real数据集可以在https://research.seas.ucla.edu/mobility-lab/v2v4real/上找到。
UDC-VIT: A Real-World Video Dataset for Under-Display Cameras
Dept. of Data Science, Seoul National University(数据科学系,首尔国立大学);Dept. of Computer Science and Engineering, Seoul National University(计算机科学与工程系,首尔国立大学);Research Center, Samsung Display Co., Ltd.(三星显示公司研究部);Mobile Display Electronics Development Team, Samsung Display Co., Ltd.(三星显示公司移动显示电子开发团队)